Instructions to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

AutomatosX commited on 16 days ago

Commit

d339afd

verified ·

1 Parent(s): de13b9b

Promote audited AXQuant v2 artifact at stable repository name

Browse files

Preserve backward-compatible repository identity; v1 remains tagged legacy-pre-v2 and this exact revision is tagged v2 after verification.

Files changed (16) hide show

LICENSE +0 -202
README.md +23 -23
axquant_manifest.json +40 -50
axquant_mtp_sidecar_manifest.json +1 -1
axquant_plan.json +0 -0
axquant_quantizer_execution.json +117 -137
axquant_runtime.json +1 -1
axquant_vision_sidecar_manifest.json +1 -1
config.json +230 -250
model-00001-of-00005.safetensors +2 -2
model-00002-of-00005.safetensors +2 -2
model-00003-of-00005.safetensors +2 -2
model-00004-of-00005.safetensors +2 -2
model-00005-of-00005.safetensors +2 -2
model-manifest.json +0 -0
model.safetensors.index.json +1 -5

LICENSE DELETED Viewed

@@ -1,202 +0,0 @@
-                                 Apache License
-                           Version 2.0, January 2004
-                        http://www.apache.org/licenses/
-   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
-   1. Definitions.
-      "License" shall mean the terms and conditions for use, reproduction,
-      and distribution as defined by Sections 1 through 9 of this document.
-      "Licensor" shall mean the copyright owner or entity authorized by
-      the copyright owner that is granting the License.
-      "Legal Entity" shall mean the union of the acting entity and all
-      other entities that control, are controlled by, or are under common
-      control with that entity. For the purposes of this definition,
-      "control" means (i) the power, direct or indirect, to cause the
-      direction or management of such entity, whether by contract or
-      otherwise, or (ii) ownership of fifty percent (50%) or more of the
-      outstanding shares, or (iii) beneficial ownership of such entity.
-      "You" (or "Your") shall mean an individual or Legal Entity
-      exercising permissions granted by this License.
-      "Source" form shall mean the preferred form for making modifications,
-      including but not limited to software source code, documentation
-      source, and configuration files.
-      "Object" form shall mean any form resulting from mechanical
-      transformation or translation of a Source form, including but
-      not limited to compiled object code, generated documentation,
-      and conversions to other media types.
-      "Work" shall mean the work of authorship, whether in Source or
-      Object form, made available under the License, as indicated by a
-      copyright notice that is included in or attached to the work
-      (an example is provided in the Appendix below).
-      "Derivative Works" shall mean any work, whether in Source or Object
-      form, that is based on (or derived from) the Work and for which the
-      editorial revisions, annotations, elaborations, or other modifications
-      represent, as a whole, an original work of authorship. For the purposes
-      of this License, Derivative Works shall not include works that remain
-      separable from, or merely link (or bind by name) to the interfaces of,
-      the Work and Derivative Works thereof.
-      "Contribution" shall mean any work of authorship, including
-      the original version of the Work and any modifications or additions
-      to that Work or Derivative Works thereof, that is intentionally
-      submitted to Licensor for inclusion in the Work by the copyright owner
-      or by an individual or Legal Entity authorized to submit on behalf of
-      the copyright owner. For the purposes of this definition, "submitted"
-      means any form of electronic, verbal, or written communication sent
-      to the Licensor or its representatives, including but not limited to
-      communication on electronic mailing lists, source code control systems,
-      and issue tracking systems that are managed by, or on behalf of, the
-      Licensor for the purpose of discussing and improving the Work, but
-      excluding communication that is conspicuously marked or otherwise
-      designated in writing by the copyright owner as "Not a Contribution."
-      "Contributor" shall mean Licensor and any individual or Legal Entity
-      on behalf of whom a Contribution has been received by Licensor and
-      subsequently incorporated within the Work.
-   2. Grant of Copyright License. Subject to the terms and conditions of
-      this License, each Contributor hereby grants to You a perpetual,
-      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
-      copyright license to reproduce, prepare Derivative Works of,
-      publicly display, publicly perform, sublicense, and distribute the
-      Work and such Derivative Works in Source or Object form.
-   3. Grant of Patent License. Subject to the terms and conditions of
-      this License, each Contributor hereby grants to You a perpetual,
-      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
-      (except as stated in this section) patent license to make, have made,
-      use, offer to sell, sell, import, and otherwise transfer the Work,
-      where such license applies only to those patent claims licensable
-      by such Contributor that are necessarily infringed by their
-      Contribution(s) alone or by combination of their Contribution(s)
-      with the Work to which such Contribution(s) was submitted. If You
-      institute patent litigation against any entity (including a
-      cross-claim or counterclaim in a lawsuit) alleging that the Work
-      or a Contribution incorporated within the Work constitutes direct
-      or contributory patent infringement, then any patent licenses
-      granted to You under this License for that Work shall terminate
-      as of the date such litigation is filed.
-   4. Redistribution. You may reproduce and distribute copies of the
-      Work or Derivative Works thereof in any medium, with or without
-      modifications, and in Source or Object form, provided that You
-      meet the following conditions:
-      (a) You must give any other recipients of the Work or
-          Derivative Works a copy of this License; and
-      (b) You must cause any modified files to carry prominent notices
-          stating that You changed the files; and
-      (c) You must retain, in the Source form of any Derivative Works
-          that You distribute, all copyright, patent, trademark, and
-          attribution notices from the Source form of the Work,
-          excluding those notices that do not pertain to any part of
-          the Derivative Works; and
-      (d) If the Work includes a "NOTICE" text file as part of its
-          distribution, then any Derivative Works that You distribute must
-          include a readable copy of the attribution notices contained
-          within such NOTICE file, excluding those notices that do not
-          pertain to any part of the Derivative Works, in at least one
-          of the following places: within a NOTICE text file distributed
-          as part of the Derivative Works; within the Source form or
-          documentation, if provided along with the Derivative Works; or,
-          within a display generated by the Derivative Works, if and
-          wherever such third-party notices normally appear. The contents
-          of the NOTICE file are for informational purposes only and
-          do not modify the License. You may add Your own attribution
-          notices within Derivative Works that You distribute, alongside
-          or as an addendum to the NOTICE text from the Work, provided
-          that such additional attribution notices cannot be construed
-          as modifying the License.
-      You may add Your own copyright statement to Your modifications and
-      may provide additional or different license terms and conditions
-      for use, reproduction, or distribution of Your modifications, or
-      for any such Derivative Works as a whole, provided Your use,
-      reproduction, and distribution of the Work otherwise complies with
-      the conditions stated in this License.
-   5. Submission of Contributions. Unless You explicitly state otherwise,
-      any Contribution intentionally submitted for inclusion in the Work
-      by You to the Licensor shall be under the terms and conditions of
-      this License, without any additional terms or conditions.
-      Notwithstanding the above, nothing herein shall supersede or modify
-      the terms of any separate license agreement you may have executed
-      with Licensor regarding such Contributions.
-   6. Trademarks. This License does not grant permission to use the trade
-      names, trademarks, service marks, or product names of the Licensor,
-      except as required for reasonable and customary use in describing the
-      origin of the Work and reproducing the content of the NOTICE file.
-   7. Disclaimer of Warranty. Unless required by applicable law or
-      agreed to in writing, Licensor provides the Work (and each
-      Contributor provides its Contributions) on an "AS IS" BASIS,
-      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
-      implied, including, without limitation, any warranties or conditions
-      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
-      PARTICULAR PURPOSE. You are solely responsible for determining the
-      appropriateness of using or redistributing the Work and assume any
-      risks associated with Your exercise of permissions under this License.
-   8. Limitation of Liability. In no event and under no legal theory,
-      whether in tort (including negligence), contract, or otherwise,
-      unless required by applicable law (such as deliberate and grossly
-      negligent acts) or agreed to in writing, shall any Contributor be
-      liable to You for damages, including any direct, indirect, special,
-      incidental, or consequential damages of any character arising as a
-      result of this License or out of the use or inability to use the
-      Work (including but not limited to damages for loss of goodwill,
-      work stoppage, computer failure or malfunction, or any and all
-      other commercial damages or losses), even if such Contributor
-      has been advised of the possibility of such damages.
-   9. Accepting Warranty or Additional Liability. While redistributing
-      the Work or Derivative Works thereof, You may choose to offer,
-      and charge a fee for, acceptance of support, warranty, indemnity,
-      or other liability obligations and/or rights consistent with this
-      License. However, in accepting such obligations, You may act only
-      on Your own behalf and on Your sole responsibility, not on behalf
-      of any other Contributor, and only if You agree to indemnify,
-      defend, and hold each Contributor harmless for any liability
-      incurred by, or claims asserted against, such Contributor by reason
-      of your accepting any such warranty or additional liability.
-   END OF TERMS AND CONDITIONS
-   APPENDIX: How to apply the Apache License to your work.
-      To apply the Apache License to your work, attach the following
-      boilerplate notice, with the fields enclosed by brackets "[]"
-      replaced with your own identifying information. (Don't include
-      the brackets!)  The text should be enclosed in the appropriate
-      comment syntax for the file format. We also recommend that a
-      file or class name and description of purpose be included on the
-      same "printed page" as the copyright notice for easier
-      identification within third-party archives.
-   Copyright 2026 Alibaba Cloud
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
-       http://www.apache.org/licenses/LICENSE-2.0
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License.

README.md CHANGED Viewed

@@ -15,6 +15,7 @@ tags:
 - qwen3.6
 - 6bit
 - 6-bit
 - mtp
 - vision
 ---
@@ -28,6 +29,9 @@ the BF16 source model. The language path is quantized while the multi-token-pred
 > artifact-integrity records, but it does not publish measured quality, long-context, kernel-speed,
 > or MTP-speed evidence. Do not interpret the AXQ product label as a benchmark claim.
 ## Model details
 | Property | Value |
@@ -37,17 +41,18 @@ the BF16 source model. The language path is quantized while the multi-token-pred
 | Product family | `qwen3.6` |
 | Source architecture | `Qwen3_5MoeForConditionalGeneration` (mixture of experts (MoE)); text path optimized |
 | Main-model parameters | 35.11B logical parameters |
-| Quantizer | AXQuant `1.0.0` |
 | Hub budget class | `6bit` |
-| AXQuant base precision class | `4bit` |
 | Planned storage-adjusted BPW | 6.0000 |
 | Measured main-model BPW | 5.7595 |
 | Measured total BPW, including MTP | **6.0001** |
 | Safetensors weight size | 26.96 GB |
 | Approximate complete download | 26.99 GB |
 | Configured maximum context | 262,144 tokens; practical limits depend on unified memory |
-| Primary runtime | AX Engine, compatibility level A |
-| Compatible runtime | MLX-LM standard text inference, compatibility level B |
 | MTP present | `True` |
 | Vision sidecar present | `True` |
@@ -65,7 +70,7 @@ protected models.
 | Sibling | Intended trade-off |
 | --- | --- |
 | [4bit sibling](https://huggingface.co/AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-4bit-MTP) | Lower-storage AXQ budget; check its exact BPW |
-| [6bit sibling](https://huggingface.co/AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP) | Higher average precision near a 6-BPW budget |
 See the [AutomatosX MLX model catalog](https://huggingface.co/collections/AutomatosX/automatosx-mlx-model-catalog)
 for related MLX and OptiQ alternatives.
@@ -96,19 +101,13 @@ metadata and optional sidecars (`vision.safetensors`, `mtp.safetensors`); this c
 does not establish MTP acceleration or vision-language quality. The artifact records MLX
 `0.32.0` and MLX-LM `0.31.3` from conversion.
-## Serve with AX Engine and MTP
-After installing [AX Engine](https://github.com/defai-digital/ax-engine), download the complete
-repository and serve the local directory:
-```bash
-ax-engine serve ./AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP --port 31418
-```
-AX Engine is the authority for the AXQ runtime contract and native MTP sidecar.
-This development package does not claim runtime speedups until identical-checkpoint benchmarks are
-published. The artifact records AX Engine version `not recorded`. Native
-`model-manifest.json` status: included as `model-manifest.json`.
 ## Quantization layout
@@ -116,7 +115,7 @@ published. The artifact records AX Engine version `not recorded`. Native
 | --- | ---: | ---: |
 | `4bit` | 24.70B | 68.69% |
 | `6bit` | 8.75B | 24.35% |
-| `8bit` | 701.96M | 1.95% |
 | `bf16` | 1.80B | 5.01% |
 - Quantization methods: `affine, bf16`.
@@ -135,8 +134,8 @@ establish MTP acceleration or vision-language quality.
 | --- | --- |
 | Planning evidence | `architecture_prior` |
 | Calibration | none; the allocation is based on architecture priors |
-| Quantizer execution | 471/471 recorded module conversions succeeded; 0 fallbacks |
-| AX Engine native manifest | included as `model-manifest.json` |
 | Quality versus BF16 or uniform baselines | Not published; no quality-retention claim |
 | MTP acceptance and speed | not measured; no MTP speedup claim |
 | AX Engine kernel evidence | `unmeasured` |
@@ -154,6 +153,8 @@ establish MTP acceleration or vision-language quality.
 - MTP may be ignored outside AX Engine and its speedup is unmeasured for this exact checkpoint.
 - Vision weights are byte-preserved at BF16, but this release does not claim validated VLM quality.
 - The configured context window can require substantially more memory as the KV cache grows.
 - Upstream capabilities, limitations, biases, and responsible-use guidance still apply.
 ## Provenance and audit files
@@ -163,10 +164,9 @@ establish MTP acceleration or vision-language quality.
 - [`axquant_plan.json`](axquant_plan.json): per-tensor precision decisions and planning evidence.
 - [`axquant_quantizer_execution.json`](axquant_quantizer_execution.json): conversion coverage and
   fallback records.
-- [`axquant_runtime.json`](axquant_runtime.json): AX Engine and MLX-LM compatibility contract.
 - [`axquant_mtp_sidecar_manifest.json`](axquant_mtp_sidecar_manifest.json): MTP tensor provenance.
 - [`axquant_vision_sidecar_manifest.json`](axquant_vision_sidecar_manifest.json): protected vision tensor provenance.
-- [`model-manifest.json`](model-manifest.json): AX Engine native tensor manifest.
 All published provenance uses repository-relative paths. Local source paths are stripped before
 publication. The checkpoint was converted from BF16 rather than re-quantized from an OptiQ
@@ -176,5 +176,5 @@ identical BPW or quality.
 ## License
 The checkpoint follows the upstream model license where applicable (often Apache License 2.0). See
-[`LICENSE`](LICENSE) and the [Qwen/Qwen3.6-35B-A3B model card](https://huggingface.co/Qwen/Qwen3.6-35B-A3B/tree/995ad96eacd98c81ed38be0c5b274b04031597b0) for license terms, model
 limitations, and responsible-use guidance.

 - qwen3.6
 - 6bit
 - 6-bit
+- v2
 - mtp
 - vision
 ---
 > artifact-integrity records, but it does not publish measured quality, long-context, kernel-speed,
 > or MTP-speed evidence. Do not interpret the AXQ product label as a benchmark claim.
+> **Stable-name v2.** `main` serves the audited v2 artifact for backward compatibility. The same revision is tagged `v2`; the replaced artifact remains recoverable at `legacy-pre-v2`.
 ## Model details
 | Property | Value |
 | Product family | `qwen3.6` |
 | Source architecture | `Qwen3_5MoeForConditionalGeneration` (mixture of experts (MoE)); text path optimized |
 | Main-model parameters | 35.11B logical parameters |
+| Quantizer | AXQuant `1.2.0` |
 | Hub budget class | `6bit` |
+| Artifact edition | `v2` |
+| AXQuant base precision class | `6bit` |
 | Planned storage-adjusted BPW | 6.0000 |
 | Measured main-model BPW | 5.7595 |
 | Measured total BPW, including MTP | **6.0001** |
 | Safetensors weight size | 26.96 GB |
 | Approximate complete download | 26.99 GB |
 | Configured maximum context | 262,144 tokens; practical limits depend on unified memory |
+| MLX-LM compatibility | Standard text inference, compatibility level B |
+| AX Engine native execution | Not established; no validated native manifest is included |
 | MTP present | `True` |
 | Vision sidecar present | `True` |
 | Sibling | Intended trade-off |
 | --- | --- |
 | [4bit sibling](https://huggingface.co/AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-4bit-MTP) | Lower-storage AXQ budget; check its exact BPW |
+| [6bit sibling](https://huggingface.co/AutomatosX/AX-Qwen3.6-35B-A3B-MLX-AXQ-6bit-MTP) | Higher average precision near the 6-BPW budget |
 See the [AutomatosX MLX model catalog](https://huggingface.co/collections/AutomatosX/automatosx-mlx-model-catalog)
 for related MLX and OptiQ alternatives.
 does not establish MTP acceleration or vision-language quality. The artifact records MLX
 `0.32.0` and MLX-LM `0.31.3` from conversion.
+## AX Engine status
+This package does **not** include a validated native `model-manifest.json`, so AX Engine execution
+is not established by this release. The AX Engine fields in `axquant_runtime.json` describe the
+intended compatibility contract, not observed runtime evidence. Use the MLX-LM path above for
+standard text/backbone inference. The artifact records AX Engine version
+`not recorded`, but version discovery alone is not a runtime check.
 ## Quantization layout
 | --- | ---: | ---: |
 | `4bit` | 24.70B | 68.69% |
 | `6bit` | 8.75B | 24.35% |
+| `8bit` | 701.90M | 1.95% |
 | `bf16` | 1.80B | 5.01% |
 - Quantization methods: `affine, bf16`.
 | --- | --- |
 | Planning evidence | `architecture_prior` |
 | Calibration | none; the allocation is based on architecture priors |
+| Quantizer execution | 469/469 recorded module conversions succeeded; 0 fallbacks |
+| AX Engine native manifest | not included |
 | Quality versus BF16 or uniform baselines | Not published; no quality-retention claim |
 | MTP acceptance and speed | not measured; no MTP speedup claim |
 | AX Engine kernel evidence | `unmeasured` |
 - MTP may be ignored outside AX Engine and its speedup is unmeasured for this exact checkpoint.
 - Vision weights are byte-preserved at BF16, but this release does not claim validated VLM quality.
 - The configured context window can require substantially more memory as the KV cache grows.
+- AX Engine execution is not established because this package has no validated native manifest.
 - Upstream capabilities, limitations, biases, and responsible-use guidance still apply.
 ## Provenance and audit files
 - [`axquant_plan.json`](axquant_plan.json): per-tensor precision decisions and planning evidence.
 - [`axquant_quantizer_execution.json`](axquant_quantizer_execution.json): conversion coverage and
   fallback records.
+- [`axquant_runtime.json`](axquant_runtime.json): declared AX Engine and MLX-LM compatibility metadata; runtime checks remain separate evidence.
 - [`axquant_mtp_sidecar_manifest.json`](axquant_mtp_sidecar_manifest.json): MTP tensor provenance.
 - [`axquant_vision_sidecar_manifest.json`](axquant_vision_sidecar_manifest.json): protected vision tensor provenance.
 All published provenance uses repository-relative paths. Local source paths are stripped before
 publication. The checkpoint was converted from BF16 rather than re-quantized from an OptiQ
 ## License
 The checkpoint follows the upstream model license where applicable (often Apache License 2.0). See
+the [Qwen/Qwen3.6-35B-A3B model card](https://huggingface.co/Qwen/Qwen3.6-35B-A3B/tree/995ad96eacd98c81ed38be0c5b274b04031597b0) for license terms, model
 limitations, and responsible-use guidance.

axquant_manifest.json CHANGED Viewed

@@ -1,42 +1,37 @@
 {
-  "axquant_version": "1.0.0",
   "calibration": null,
-  "created_at": "2026-08-02T04:43:16.003828Z",
-  "effective_bpw": 5.999999988428959,
   "files": [
-    {
-      "path": "LICENSE",
-      "sha256": "50cbab8a892c5f2993b8c7351a99182507472def3b1374558308605d99b86b32",
-      "size_bytes": 11343
-    },
     {
       "path": "README.md",
-      "sha256": "7d29cc018eaa96449cbc0612c8e813ded18f9fd0583fbc942464967afadd42a2",
-      "size_bytes": 8161
     },
     {
       "path": "axquant_mtp_sidecar_manifest.json",
-      "sha256": "72bb9b4c7769171e3c6deb53b77b62b77101f205c352b4022337aa22a67a1da7",
       "size_bytes": 1066
     },
     {
       "path": "axquant_plan.json",
-      "sha256": "79021f1bf83cb4fc7a31e7faa1ea2b1912030a3a674f7a8b60c8365efd8fd0ba",
-      "size_bytes": 990979
     },
     {
       "path": "axquant_quantizer_execution.json",
-      "sha256": "c82b680a3723b1039ca91bccdee66aa25a6b4f6699789c6701bf47abb4f180d3",
-      "size_bytes": 124996
     },
     {
       "path": "axquant_runtime.json",
-      "sha256": "8bbe1ed5112b4a2304919332f7132344a95a8c529ffeab905474cd5a718c9fa1",
       "size_bytes": 1676
     },
     {
       "path": "axquant_vision_sidecar_manifest.json",
-      "sha256": "0872a80a95812d977576655e02df0ea600e5b0fda3d35e23993682106154e41c",
       "size_bytes": 1072
     },
     {
@@ -46,8 +41,8 @@
     },
     {
       "path": "config.json",
-      "sha256": "49b09920aa58e16ab7b268196af5c94d16669557ea6a566b93ce4cea87651854",
-      "size_bytes": 139529
     },
     {
       "path": "generation_config.json",
@@ -56,38 +51,33 @@
     },
     {
       "path": "model-00001-of-00005.safetensors",
-      "sha256": "e85951358771ed2e1052efc85364cafb359f0f465cd7674ae4e0bbd854347003",
-      "size_bytes": 5367595602
     },
     {
       "path": "model-00002-of-00005.safetensors",
-      "sha256": "5d6a0c566e70232cc29ceacdd17b21054166f24ac9de919d2bb571ad824b1222",
-      "size_bytes": 5358169404
     },
     {
       "path": "model-00003-of-00005.safetensors",
-      "sha256": "72c7e4ddc900bacab7c7ca7ad7db6c5795df3d626147b30516dcdf79ea1cd3fc",
-      "size_bytes": 5348491760
     },
     {
       "path": "model-00004-of-00005.safetensors",
-      "sha256": "9e77ebe4d3dca3f5bb296bcd0f4a5cbf7d38156c51b46fb13aeca46e68ad7419",
-      "size_bytes": 5253722500
     },
     {
       "path": "model-00005-of-00005.safetensors",
-      "sha256": "d0b612e38f52e9ff1909ae2c54d69e0ae486df7ea95aa003477abba7bb6e53dc",
-      "size_bytes": 3053700049
-    },
-    {
-      "path": "model-manifest.json",
-      "sha256": "8b624295c531a13d88e1284c4c81889e9b3704f11234b5393c0aa650f8ddddf7",
-      "size_bytes": 315982
     },
     {
       "path": "model.safetensors.index.json",
-      "sha256": "e7828224d66e3682de34154c599082c621f546fb71679f8c5b386a19318696b4",
-      "size_bytes": 187109
     },
     {
       "path": "mtp.safetensors",
@@ -118,9 +108,9 @@
   "format": "mlx",
   "logical_parameters": 35951822704,
   "main_logical_parameters": 35107181936,
-  "main_weight_file_size_bytes": 25274858939,
-  "measured_main_bpw": 5.75947314371761,
-  "measured_total_bpw": 6.000061345874399,
   "mtp_acceptance_retention": null,
   "mtp_distribution": {
     "bf16": {
@@ -143,7 +133,7 @@
   },
   "mtp_present": true,
   "mtp_weight_file_size_bytes": 1689283776,
-  "plan_sha256": "b907f067d21ad6833d8d003727dee5bf475ca74c10498c7986025e40e72b0940",
   "profile": "general",
   "protected_weight_file_size_bytes": 893179624,
   "quantizer": "axquant",
@@ -170,7 +160,7 @@
         "support_level": "standard-inference"
       }
     ],
-    "created_at": "2026-08-02T04:43:03.045367Z",
     "kv_cache": null,
     "memory_policy": {
       "kv_cache_precision": "runtime-default",
@@ -208,7 +198,7 @@
   "schema_version": "axquant.artifact.v2",
   "software_versions": {
     "ax_engine": null,
-    "axquant": "1.0.0",
     "mlx": "0.32.0",
     "mlx_lm": "0.31.3",
     "pydantic": "2.13.4",
@@ -222,24 +212,24 @@
     "model_id": "Qwen/Qwen3.6-35B-A3B",
     "revision": "995ad96eacd98c81ed38be0c5b274b04031597b0"
   },
-  "target_class": "4bit",
   "weight_distribution": {
     "4bit": {
       "fraction": 0.6869209985632332,
       "parameters": 24696061952
     },
     "6bit": {
-      "fraction": 0.2434611881590681,
-      "parameters": 8752873472
     },
     "8bit": {
-      "fraction": 0.019524909815543244,
-      "parameters": 701956096
     },
     "bf16": {
-      "fraction": 0.05009290346215543,
-      "parameters": 1800931184
     }
   },
-  "weight_file_size_bytes": 26964142715
 }

 {
+  "axquant_version": "1.2.0",
   "calibration": null,
+  "created_at": "2026-08-05T13:39:34.268276Z",
+  "effective_bpw": 5.999999959946398,
   "files": [
     {
       "path": "README.md",
+      "sha256": "41cee394e2a7e2344ad3d975df45e76ac70f89b836ebc1130ef692ec2fe2c724",
+      "size_bytes": 8314
     },
     {
       "path": "axquant_mtp_sidecar_manifest.json",
+      "sha256": "c0f5463aa1557c67bd0adb7498e507c9dffe8de34aadc0f72a1fd86a3ba2bcf7",
       "size_bytes": 1066
     },
     {
       "path": "axquant_plan.json",
+      "sha256": "07b6cecbf58a9f1c71814ba8b9050bf0aff2adec41ecc4a884e18629635b01cf",
+      "size_bytes": 987594
     },
     {
       "path": "axquant_quantizer_execution.json",
+      "sha256": "c164ccc95fcb87945b85399defcfd22dd132f7426cffd404ebf7533f22e855bb",
+      "size_bytes": 124464
     },
     {
       "path": "axquant_runtime.json",
+      "sha256": "4c7df1a009cc989f9dedd9113b377c88e52235e55dbd209b1ab5466eda033620",
       "size_bytes": 1676
     },
     {
       "path": "axquant_vision_sidecar_manifest.json",
+      "sha256": "af9e491342387394a21292114a19b464357375314c30b1799b168d3120c1e84d",
       "size_bytes": 1072
     },
     {
     },
     {
       "path": "config.json",
+      "sha256": "30898a68ce3c1dc40ef83ae9ca8606013e97e1a6e9281409c9351395e4ff03ba",
+      "size_bytes": 138993
     },
     {
       "path": "generation_config.json",
     },
     {
       "path": "model-00001-of-00005.safetensors",
+      "sha256": "860f667408f32e2353c15e7f88efcb8882b5e93e73afc222cb29b9cdf3a89c44",
+      "size_bytes": 5367600594
     },
     {
       "path": "model-00002-of-00005.safetensors",
+      "sha256": "2dba4e082f746e20d94ee9623e5ba879379a1c2a9079bcf79a457df5b8856e29",
+      "size_bytes": 5358260540
     },
     {
       "path": "model-00003-of-00005.safetensors",
+      "sha256": "f6c517b13633c3c3455f8c1f05b7a804fa0e6bcecb091f61a2c7cb825436e8ba",
+      "size_bytes": 5349479920
     },
     {
       "path": "model-00004-of-00005.safetensors",
+      "sha256": "d2fbee3ba144584f68333788d2d31f0339a51d4aa6f0d5cbfdfd8c5260b0333c",
+      "size_bytes": 5252632964
     },
     {
       "path": "model-00005-of-00005.safetensors",
+      "sha256": "860cfe8a94c0b2d8ca4ab509545096e5af73b480724147e1ffc92e2ce02b0df0",
+      "size_bytes": 3053704637
     },
     {
       "path": "model.safetensors.index.json",
+      "sha256": "e2316c3702cf36cd3ed4a1ce3aa9cd507a3832944b3e8e99a1b10258da701e92",
+      "size_bytes": 186677
     },
     {
       "path": "mtp.safetensors",
   "format": "mlx",
   "logical_parameters": 35951822704,
   "main_logical_parameters": 35107181936,
+  "main_weight_file_size_bytes": 25274858279,
+  "measured_main_bpw": 5.759472993321032,
+  "measured_total_bpw": 6.000061199011191,
   "mtp_acceptance_retention": null,
   "mtp_distribution": {
     "bf16": {
   },
   "mtp_present": true,
   "mtp_weight_file_size_bytes": 1689283776,
+  "plan_sha256": "b1f023ba3430624efdd56327af37b5b814d093e4b9a114788d5c1dcc1f88da3c",
   "profile": "general",
   "protected_weight_file_size_bytes": 893179624,
   "quantizer": "axquant",
         "support_level": "standard-inference"
       }
     ],
+    "created_at": "2026-08-05T13:39:05.854633Z",
     "kv_cache": null,
     "memory_policy": {
       "kv_cache_precision": "runtime-default",
   "schema_version": "axquant.artifact.v2",
   "software_versions": {
     "ax_engine": null,
+    "axquant": "1.2.0",
     "mlx": "0.32.0",
     "mlx_lm": "0.31.3",
     "pydantic": "2.13.4",
     "model_id": "Qwen/Qwen3.6-35B-A3B",
     "revision": "995ad96eacd98c81ed38be0c5b274b04031597b0"
   },
+  "target_class": "6bit",
   "weight_distribution": {
     "4bit": {
       "fraction": 0.6869209985632332,
       "parameters": 24696061952
     },
     "6bit": {
+      "fraction": 0.2434625553220185,
+      "parameters": 8752922624
     },
     "8bit": {
+      "fraction": 0.019523428722346983,
+      "parameters": 701902848
     },
     "bf16": {
+      "fraction": 0.0500930173924013,
+      "parameters": 1800935280
     }
   },
+  "weight_file_size_bytes": 26964142055
 }

axquant_mtp_sidecar_manifest.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "created_at": "2026-08-02T04:43:03.034795Z",
   "dtypes": [
     "BF16"
   ],

 {
+  "created_at": "2026-08-05T13:39:05.842150Z",
   "dtypes": [
     "BF16"
   ],

axquant_plan.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

axquant_quantizer_execution.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
-  "created_at": "2026-08-02T04:42:55.190223Z",
-  "plan_sha256": "b907f067d21ad6833d8d003727dee5bf475ca74c10498c7986025e40e72b0940",
   "records": [
     {
       "bits": 8,
@@ -173,9 +173,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.0.mlp.shared_expert_gate",
@@ -243,9 +243,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.1.mlp.shared_expert_gate",
@@ -363,9 +363,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.2.mlp.shared_expert_gate",
@@ -433,7 +433,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -593,7 +593,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -713,7 +713,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -833,7 +833,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -903,7 +903,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1163,7 +1163,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1233,7 +1233,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1303,7 +1303,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1503,7 +1503,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1583,7 +1583,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1653,7 +1653,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1813,9 +1813,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.14.linear_attn.in_proj_b",
@@ -1863,7 +1863,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1923,7 +1923,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -1933,9 +1933,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.16.linear_attn.in_proj_a",
@@ -1943,9 +1943,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.16.linear_attn.in_proj_b",
@@ -2003,7 +2003,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2153,9 +2153,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.17.linear_attn.in_proj_a",
@@ -2163,9 +2163,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.17.linear_attn.in_proj_b",
@@ -2213,7 +2213,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2223,9 +2223,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.18.linear_attn.in_proj_a",
@@ -2233,9 +2233,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.18.linear_attn.in_proj_b",
@@ -2293,7 +2293,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2353,7 +2353,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2493,9 +2493,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.20.linear_attn.in_proj_a",
@@ -2503,9 +2503,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.20.linear_attn.in_proj_b",
@@ -2553,7 +2553,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2563,9 +2563,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.21.linear_attn.in_proj_a",
@@ -2573,9 +2573,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.21.linear_attn.in_proj_b",
@@ -2633,7 +2633,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2643,9 +2643,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.22.linear_attn.in_proj_a",
@@ -2653,9 +2653,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.22.linear_attn.in_proj_b",
@@ -2713,7 +2713,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2743,9 +2743,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.23.self_attn.k_proj",
@@ -2773,9 +2773,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.23.self_attn.v_proj",
@@ -2913,7 +2913,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2923,9 +2923,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.24.linear_attn.in_proj_a",
@@ -2933,9 +2933,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.24.linear_attn.in_proj_b",
@@ -2983,7 +2983,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -2993,9 +2993,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.25.linear_attn.in_proj_a",
@@ -3003,9 +3003,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.25.linear_attn.in_proj_b",
@@ -3063,7 +3063,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3123,9 +3123,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.27.self_attn.k_proj",
@@ -3153,9 +3153,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.27.self_attn.v_proj",
@@ -3213,9 +3213,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.26.linear_attn.in_proj_a",
@@ -3223,9 +3223,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.26.linear_attn.in_proj_b",
@@ -3283,7 +3283,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3343,7 +3343,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3353,9 +3353,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.28.linear_attn.in_proj_a",
@@ -3363,9 +3363,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.28.linear_attn.in_proj_b",
@@ -3413,7 +3413,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3523,9 +3523,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.31.self_attn.k_proj",
@@ -3553,9 +3553,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.31.self_attn.v_proj",
@@ -3563,9 +3563,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.29.linear_attn.in_proj_a",
@@ -3573,9 +3573,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.29.linear_attn.in_proj_b",
@@ -3633,7 +3633,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3643,9 +3643,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.30.linear_attn.in_proj_a",
@@ -3653,9 +3653,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.30.linear_attn.in_proj_b",
@@ -3713,7 +3713,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3763,7 +3763,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3773,9 +3773,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.32.linear_attn.in_proj_a",
@@ -3783,9 +3783,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.32.linear_attn.in_proj_b",
@@ -3883,7 +3883,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -3893,9 +3893,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.33.linear_attn.in_proj_a",
@@ -3903,9 +3903,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.33.linear_attn.in_proj_b",
@@ -4003,7 +4003,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -4013,9 +4013,9 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.34.linear_attn.in_proj_a",
@@ -4043,9 +4043,9 @@
       "success": true
     },
     {
-      "bits": 8,
       "fallback": false,
-      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.34.linear_attn.in_proj_z",
@@ -4123,7 +4123,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -4193,7 +4193,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -4353,7 +4353,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -4473,7 +4473,7 @@
       "success": true
     },
     {
-      "bits": 6,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
@@ -4592,16 +4592,6 @@
       "note": "MLX-LM affine packing",
       "success": true
     },
-    {
-      "bits": 6,
-      "fallback": false,
-      "group_size": 32,
-      "metadata": {},
-      "method": "affine",
-      "module_path": "model.language_model.layers.38.mlp.shared_expert_gate",
-      "note": "MLX-LM affine packing",
-      "success": true
-    },
     {
       "bits": 6,
       "fallback": false,
@@ -4662,16 +4652,6 @@
       "note": "MLX-LM affine packing",
       "success": true
     },
-    {
-      "bits": 6,
-      "fallback": false,
-      "group_size": 32,
-      "metadata": {},
-      "method": "affine",
-      "module_path": "model.language_model.layers.39.mlp.shared_expert_gate",
-      "note": "MLX-LM affine packing",
-      "success": true
-    },
     {
       "bits": 8,
       "fallback": false,

 {
+  "created_at": "2026-08-05T13:38:58.824011Z",
+  "plan_sha256": "b1f023ba3430624efdd56327af37b5b814d093e4b9a114788d5c1dcc1f88da3c",
   "records": [
     {
       "bits": 8,
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.0.mlp.shared_expert_gate",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.1.mlp.shared_expert_gate",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.2.mlp.shared_expert_gate",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.14.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.16.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.16.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.17.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.17.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.18.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.18.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.20.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.20.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.21.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.21.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.22.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.22.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.23.self_attn.k_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.23.self_attn.v_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.24.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.24.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.25.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.25.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.27.self_attn.k_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.27.self_attn.v_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.26.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.26.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.28.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.28.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.31.self_attn.k_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.31.self_attn.v_proj",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.29.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.29.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.30.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.30.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.32.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.32.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.33.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.33.linear_attn.in_proj_b",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.34.linear_attn.in_proj_a",
       "success": true
     },
     {
+      "bits": 6,
       "fallback": false,
+      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "model.language_model.layers.34.linear_attn.in_proj_z",
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "success": true
     },
     {
+      "bits": 8,
       "fallback": false,
       "group_size": 32,
       "metadata": {},
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 6,
       "fallback": false,
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 8,
       "fallback": false,

axquant_runtime.json CHANGED Viewed

@@ -21,7 +21,7 @@
       "support_level": "standard-inference"
     }
   ],
-  "created_at": "2026-08-02T04:43:03.045367Z",
   "kv_cache": null,
   "memory_policy": {
     "kv_cache_precision": "runtime-default",

       "support_level": "standard-inference"
     }
   ],
+  "created_at": "2026-08-05T13:39:05.854633Z",
   "kv_cache": null,
   "memory_policy": {
     "kv_cache_precision": "runtime-default",

axquant_vision_sidecar_manifest.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "created_at": "2026-08-02T04:42:58.818382Z",
   "dtypes": [
     "BF16"
   ],

 {
+  "created_at": "2026-08-05T13:39:01.669426Z",
   "dtypes": [
     "BF16"
   ],

config.json CHANGED Viewed

@@ -62,8 +62,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.switch_mlp.down_proj": {
@@ -127,8 +127,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.switch_mlp.down_proj": {
@@ -192,7 +192,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.10.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -232,7 +232,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.11.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -317,7 +317,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.12.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -382,7 +382,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.13.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -407,8 +407,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_qkv": {
@@ -447,7 +447,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.14.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -487,7 +487,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.15.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -527,13 +527,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_qkv": {
@@ -572,7 +572,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.16.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -592,13 +592,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_qkv": {
@@ -637,7 +637,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.17.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -657,13 +657,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_qkv": {
@@ -702,7 +702,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.18.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -742,7 +742,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.19.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -827,8 +827,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.switch_mlp.down_proj": {
@@ -847,13 +847,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_qkv": {
@@ -892,7 +892,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.20.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -912,13 +912,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_qkv": {
@@ -957,7 +957,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.21.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -977,13 +977,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_qkv": {
@@ -1022,7 +1022,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.22.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1062,7 +1062,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1082,8 +1082,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.o_proj": {
@@ -1097,18 +1097,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_qkv": {
@@ -1147,7 +1147,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.24.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1167,13 +1167,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_qkv": {
@@ -1212,7 +1212,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.25.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1232,13 +1232,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_qkv": {
@@ -1277,7 +1277,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.26.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1317,7 +1317,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1337,8 +1337,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.o_proj": {
@@ -1352,18 +1352,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_qkv": {
@@ -1402,7 +1402,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.28.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1422,13 +1422,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_qkv": {
@@ -1467,7 +1467,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.29.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1507,7 +1507,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.3.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1547,13 +1547,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_qkv": {
@@ -1592,7 +1592,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.30.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1632,7 +1632,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1652,8 +1652,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.o_proj": {
@@ -1667,18 +1667,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_qkv": {
@@ -1717,7 +1717,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.32.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1737,13 +1737,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_qkv": {
@@ -1782,7 +1782,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.33.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1802,8 +1802,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_b": {
@@ -1817,8 +1817,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_z": {
-      "bits": 8,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.out_proj": {
@@ -1847,7 +1847,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1887,7 +1887,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.35.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -1972,7 +1972,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.36.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2037,7 +2037,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.37.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2101,11 +2101,6 @@
       "group_size": 64,
       "mode": "affine"
     },
-    "language_model.model.layers.38.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 32,
-      "mode": "affine"
-    },
     "language_model.model.layers.38.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
@@ -2141,11 +2136,6 @@
       "group_size": 64,
       "mode": "affine"
     },
-    "language_model.model.layers.39.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 32,
-      "mode": "affine"
-    },
     "language_model.model.layers.39.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
@@ -2227,7 +2217,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.4.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2292,7 +2282,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.5.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2357,7 +2347,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.6.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2397,7 +2387,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.7.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2482,7 +2472,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.8.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2547,7 +2537,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.9.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2622,8 +2612,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.switch_mlp.down_proj": {
@@ -2687,8 +2677,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.switch_mlp.down_proj": {
@@ -2752,7 +2742,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.10.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2792,7 +2782,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.11.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2877,7 +2867,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.12.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2942,7 +2932,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.13.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -2967,8 +2957,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_qkv": {
@@ -3007,7 +2997,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.14.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3047,7 +3037,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.15.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3087,13 +3077,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_qkv": {
@@ -3132,7 +3122,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.16.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3152,13 +3142,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_qkv": {
@@ -3197,7 +3187,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.17.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3217,13 +3207,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_qkv": {
@@ -3262,7 +3252,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.18.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3302,7 +3292,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.19.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3387,8 +3377,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.switch_mlp.down_proj": {
@@ -3407,13 +3397,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_qkv": {
@@ -3452,7 +3442,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.20.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3472,13 +3462,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_qkv": {
@@ -3517,7 +3507,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.21.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3537,13 +3527,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_qkv": {
@@ -3582,7 +3572,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.22.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3622,7 +3612,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3642,8 +3632,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.o_proj": {
@@ -3657,18 +3647,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_qkv": {
@@ -3707,7 +3697,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.24.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3727,13 +3717,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_qkv": {
@@ -3772,7 +3762,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.25.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3792,13 +3782,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_qkv": {
@@ -3837,7 +3827,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.26.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3877,7 +3867,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3897,8 +3887,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.o_proj": {
@@ -3912,18 +3902,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_qkv": {
@@ -3962,7 +3952,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.28.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -3982,13 +3972,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_qkv": {
@@ -4027,7 +4017,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.29.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4067,7 +4057,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.3.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4107,13 +4097,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_qkv": {
@@ -4152,7 +4142,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.30.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4192,7 +4182,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4212,8 +4202,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.k_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.o_proj": {
@@ -4227,18 +4217,18 @@
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.v_proj": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_qkv": {
@@ -4277,7 +4267,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.32.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4297,13 +4287,13 @@
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_b": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_qkv": {
@@ -4342,7 +4332,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.33.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4362,8 +4352,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_a": {
-      "bits": 6,
-      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_b": {
@@ -4377,8 +4367,8 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_z": {
-      "bits": 8,
-      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.out_proj": {
@@ -4407,7 +4397,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.34.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4447,7 +4437,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.35.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4532,7 +4522,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.36.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4597,7 +4587,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.37.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4661,11 +4651,6 @@
       "group_size": 64,
       "mode": "affine"
     },
-    "language_model.model.layers.38.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 32,
-      "mode": "affine"
-    },
     "language_model.model.layers.38.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
@@ -4701,11 +4686,6 @@
       "group_size": 64,
       "mode": "affine"
     },
-    "language_model.model.layers.39.mlp.shared_expert_gate": {
-      "bits": 6,
-      "group_size": 32,
-      "mode": "affine"
-    },
     "language_model.model.layers.39.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
@@ -4787,7 +4767,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.4.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4852,7 +4832,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.5.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4917,7 +4897,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.6.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -4957,7 +4937,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.7.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -5042,7 +5022,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.8.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },
@@ -5107,7 +5087,7 @@
       "mode": "affine"
     },
     "language_model.model.layers.9.mlp.shared_expert_gate": {
-      "bits": 6,
       "group_size": 32,
       "mode": "affine"
     },

       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.10.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.11.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.12.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.13.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.14.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.15.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.16.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.17.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.18.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.19.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.20.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.21.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.22.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.23.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.24.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.25.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.26.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.27.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.28.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.29.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.3.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.30.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.31.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.32.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.33.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_b": {
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_z": {
+      "bits": 6,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.out_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.34.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.35.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.36.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.37.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.38.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.39.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.4.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.5.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.6.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.7.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.8.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.9.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.0.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.1.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.10.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.11.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.12.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.13.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.14.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.14.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.15.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.16.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.16.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.17.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.17.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.18.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.18.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.19.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.shared_expert_gate": {
+      "bits": 8,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.2.mlp.switch_mlp.down_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.20.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.20.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.21.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.21.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.22.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.22.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.23.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.23.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.24.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.24.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.25.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.25.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.26.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.26.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.27.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.27.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.28.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.28.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.29.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.29.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.3.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.30.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.30.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.31.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.k_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.o_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.31.self_attn.v_proj": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.32.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.32.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_b": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.33.linear_attn.in_proj_qkv": {
       "mode": "affine"
     },
     "language_model.model.layers.33.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_a": {
+      "bits": 8,
+      "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_b": {
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.in_proj_z": {
+      "bits": 6,
+      "group_size": 32,
       "mode": "affine"
     },
     "language_model.model.layers.34.linear_attn.out_proj": {
       "mode": "affine"
     },
     "language_model.model.layers.34.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.35.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.36.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.37.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.38.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.39.mlp.switch_mlp.down_proj": {
       "bits": 6,
       "group_size": 64,
       "mode": "affine"
     },
     "language_model.model.layers.4.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.5.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.6.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.7.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.8.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },
       "mode": "affine"
     },
     "language_model.model.layers.9.mlp.shared_expert_gate": {
+      "bits": 8,
       "group_size": 32,
       "mode": "affine"
     },

model-00001-of-00005.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e85951358771ed2e1052efc85364cafb359f0f465cd7674ae4e0bbd854347003
-size 5367595602

 version https://git-lfs.github.com/spec/v1
+oid sha256:860f667408f32e2353c15e7f88efcb8882b5e93e73afc222cb29b9cdf3a89c44
+size 5367600594

model-00002-of-00005.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d6a0c566e70232cc29ceacdd17b21054166f24ac9de919d2bb571ad824b1222
-size 5358169404

 version https://git-lfs.github.com/spec/v1
+oid sha256:2dba4e082f746e20d94ee9623e5ba879379a1c2a9079bcf79a457df5b8856e29
+size 5358260540

model-00003-of-00005.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72c7e4ddc900bacab7c7ca7ad7db6c5795df3d626147b30516dcdf79ea1cd3fc
-size 5348491760

 version https://git-lfs.github.com/spec/v1
+oid sha256:f6c517b13633c3c3455f8c1f05b7a804fa0e6bcecb091f61a2c7cb825436e8ba
+size 5349479920

model-00004-of-00005.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e77ebe4d3dca3f5bb296bcd0f4a5cbf7d38156c51b46fb13aeca46e68ad7419
-size 5253722500

 version https://git-lfs.github.com/spec/v1
+oid sha256:d2fbee3ba144584f68333788d2d31f0339a51d4aa6f0d5cbfdfd8c5260b0333c
+size 5252632964

model-00005-of-00005.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d0b612e38f52e9ff1909ae2c54d69e0ae486df7ea95aa003477abba7bb6e53dc
-size 3053700049

 version https://git-lfs.github.com/spec/v1
+oid sha256:860cfe8a94c0b2d8ca4ab509545096e5af73b480724147e1ffc92e2ce02b0df0
+size 3053704637

model-manifest.json DELETED Viewed

The diff for this file is too large to render. See raw diff

model.safetensors.index.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
     "metadata": {
-        "total_size": 24381442944,
         "total_parameters": 34660608768
     },
     "weight_map": {
@@ -1440,8 +1440,6 @@
         "language_model.model.layers.38.mlp.shared_expert.up_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert.up_proj.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert.up_proj.weight": "model-00005-of-00005.safetensors",
-        "language_model.model.layers.38.mlp.shared_expert_gate.biases": "model-00005-of-00005.safetensors",
-        "language_model.model.layers.38.mlp.shared_expert_gate.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert_gate.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.switch_mlp.down_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.switch_mlp.down_proj.scales": "model-00005-of-00005.safetensors",
@@ -1466,8 +1464,6 @@
         "language_model.model.layers.39.mlp.shared_expert.up_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert.up_proj.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert.up_proj.weight": "model-00005-of-00005.safetensors",
-        "language_model.model.layers.39.mlp.shared_expert_gate.biases": "model-00005-of-00005.safetensors",
-        "language_model.model.layers.39.mlp.shared_expert_gate.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert_gate.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.switch_mlp.down_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.switch_mlp.down_proj.scales": "model-00005-of-00005.safetensors",

 {
     "metadata": {
+        "total_size": 24381442816,
         "total_parameters": 34660608768
     },
     "weight_map": {
         "language_model.model.layers.38.mlp.shared_expert.up_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert.up_proj.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert.up_proj.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.shared_expert_gate.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.switch_mlp.down_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.38.mlp.switch_mlp.down_proj.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert.up_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert.up_proj.scales": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert.up_proj.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.shared_expert_gate.weight": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.switch_mlp.down_proj.biases": "model-00005-of-00005.safetensors",
         "language_model.model.layers.39.mlp.switch_mlp.down_proj.scales": "model-00005-of-00005.safetensors",