Instructions to use AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit")

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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"

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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"
# 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"

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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit 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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit"

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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit" \
  --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

669dda7

verified ·

1 Parent(s): a41e012

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 (13) hide show

README.md +40 -36
axquant_manifest.json +49 -39
axquant_plan.json +0 -0
axquant_quantizer_execution.json +137 -137
axquant_runtime.json +3 -3
axquant_vision_sidecar_manifest.json +30 -0
config.json +0 -0
model-00001-of-00002.safetensors +1 -1
model-00002-of-00002.safetensors +2 -2
model.safetensors.index.json +1 -1
tokenizer.json +2 -2
tokenizer_config.json +5 -1003
vision.safetensors +3 -0

README.md CHANGED Viewed

@@ -1,7 +1,7 @@
 ---
 license: apache-2.0
 library_name: mlx
-base_model: mistralai/Ministral-3-14B-Instruct-2512
 base_model_relation: quantized
 pipeline_tag: text-generation
 tags:
@@ -15,39 +15,45 @@ tags:
 - mistral3
 - 4bit
 - 4-bit
 ---
 # AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
 An **AXQuant (AXQ)** mixed-precision MLX checkpoint for Apple Silicon, converted directly from
-the BF16 source model. The language path is quantized under AXQuant protection floors (embeddings, norms, and other protected tensors remain higher precision).
 > **Development evidence — not a certified AXQuant release.** This package has conversion and
 > 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 |
 | --- | --- |
-| Base model | [mistralai/Ministral-3-14B-Instruct-2512](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) |
-| Source revision | `unrecorded` |
 | Product family | `mistral3` |
 | Source architecture | `Mistral3ForConditionalGeneration` (dense); text path optimized |
-| Main-model parameters | 13.51B logical parameters |
-| Quantizer | AXQuant `1.0.1` |
 | Hub budget class | `4bit` |
-| AXQuant base precision class | `4bit` |
-| Planned storage-adjusted BPW | 5.2798 |
-| Measured main-model BPW | 5.2799 |
-| Measured total BPW | **5.2799** |
-| Safetensors weight size | 8.91 GB |
-| Approximate complete download | 8.93 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 | `False` |
-| Vision sidecar present | `False` |
 This repository contains MLX Safetensors. It does **not** contain PyTorch or GGUF weights.
@@ -63,7 +69,7 @@ protected models.
 | Sibling | Intended trade-off |
 | --- | --- |
 | [4bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit) | Lower-storage AXQ budget; check its exact BPW |
-| [6bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-6bit) | 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.
@@ -75,7 +81,7 @@ python -m pip install -U huggingface_hub
 hf download AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit --local-dir ./AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
 ```
-Allow at least 8.93 GB of free disk space. Pin the resulting Hub commit in reproducible
 deployments rather than relying indefinitely on `main`.
 ## Run with MLX-LM
@@ -94,32 +100,26 @@ 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
-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-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit --port 31418
-```
-AX Engine is the authority for the AXQ runtime contract.
-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: not included.
 ## Quantization layout
 | Main-weight precision | Parameters | Share |
 | --- | ---: | ---: |
-| `4bit` | 12.16B | 90.06% |
-| `8bit` | 671.09M | 4.97% |
-| `bf16` | 671.50M | 4.97% |
 - Quantization methods: `affine, bf16`.
 - Group sizes used by quantized assignments: `32, 64`.
 - MTP sidecar: not included.
-- Vision sidecar: not included.
 - Optimization scope: `text-path`.
 - Support tier: `convertible`.
@@ -137,7 +137,7 @@ establish MTP acceleration or vision-language quality.
 | 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` |
-| Vision-language quality | Not applicable (no vision sidecar in this package) |
 | Long-context quality | 262,144-token capacity is config metadata, not a validated claim |
 | Release certification | **Not certified**; formal AXQuant M0-M8 gates are not closed |
@@ -148,7 +148,10 @@ establish MTP acceleration or vision-language quality.
   KV-cache policy, runtime buffers, and other processes using unified memory.
 - Architecture-prior allocation is not measured sensitivity. It must not be presented as measured
   model 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
@@ -158,7 +161,8 @@ 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.
 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
@@ -168,5 +172,5 @@ identical BPW or quality.
 ## License
 The checkpoint follows the upstream model license where applicable (often Apache License 2.0). See
-the [mistralai/Ministral-3-14B-Instruct-2512 model card](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) for license terms, model
 limitations, and responsible-use guidance.

 ---
 license: apache-2.0
 library_name: mlx
+base_model: mistralai/Ministral-3-14B-Instruct-2512-BF16
 base_model_relation: quantized
 pipeline_tag: text-generation
 tags:
 - mistral3
 - 4bit
 - 4-bit
+- v2
+- vision
 ---
 # AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
 An **AXQuant (AXQ)** mixed-precision MLX checkpoint for Apple Silicon, converted directly from
+the BF16 source model. The language path is quantized while the vision tower are preserved as BF16 sidecars when present.
 > **Development evidence — not a certified AXQuant release.** This package has conversion and
 > 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 |
 | --- | --- |
+| Base model | [mistralai/Ministral-3-14B-Instruct-2512-BF16](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512-BF16/tree/3cea74c1ebaf5ce5f5a2553de470e2ceab825142) |
+| Source revision | `3cea74c1ebaf5ce5f5a2553de470e2ceab825142` |
 | Product family | `mistral3` |
 | Source architecture | `Mistral3ForConditionalGeneration` (dense); text path optimized |
+| Main-model parameters | 13.95B logical parameters |
+| Quantizer | AXQuant `1.2.0` |
 | Hub budget class | `4bit` |
+| Artifact edition | `v2` |
+| AXQuant base precision class | `5p6bpw` |
+| Planned storage-adjusted BPW | 5.6099 |
+| Measured main-model BPW | 5.6100 |
+| Measured total BPW | **5.6100** |
+| Safetensors weight size | 9.78 GB |
+| Approximate complete download | 9.80 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 | `False` |
+| Vision sidecar present | `True` |
 This repository contains MLX Safetensors. It does **not** contain PyTorch or GGUF weights.
 | Sibling | Intended trade-off |
 | --- | --- |
 | [4bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit) | Lower-storage AXQ budget; check its exact BPW |
+| [6bit sibling](https://huggingface.co/AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-6bit) | 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.
 hf download AutomatosX/AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit --local-dir ./AX-Ministral-3-14B-Instruct-2512-MLX-AXQ-4bit
 ```
+Allow at least 9.80 GB of free disk space. Pin the resulting Hub commit in reproducible
 deployments rather than relying indefinitely on `main`.
 ## Run with MLX-LM
 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
 | Main-weight precision | Parameters | Share |
 | --- | ---: | ---: |
+| `4bit` | 12.16B | 87.22% |
+| `8bit` | 671.09M | 4.81% |
+| `bf16` | 1.11B | 7.96% |
 - Quantization methods: `affine, bf16`.
 - Group sizes used by quantized assignments: `32, 64`.
 - MTP sidecar: not included.
+- Vision sidecar: 222 tensors, 438.96M parameters, 0.88 GB, BF16.
 - Optimization scope: `text-path`.
 - Support tier: `convertible`.
 | 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` |
+| Vision-language quality | Not evaluated or claimed; vision tensors are preserved at BF16 |
 | Long-context quality | 262,144-token capacity is config metadata, not a validated claim |
 | Release certification | **Not certified**; formal AXQuant M0-M8 gates are not closed |
   KV-cache policy, runtime buffers, and other processes using unified memory.
 - Architecture-prior allocation is not measured sensitivity. It must not be presented as measured
   model quality.
+- 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_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 [mistralai/Ministral-3-14B-Instruct-2512-BF16 model card](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512-BF16/tree/3cea74c1ebaf5ce5f5a2553de470e2ceab825142) for license terms, model
 limitations, and responsible-use guidance.

axquant_manifest.json CHANGED Viewed

@@ -1,28 +1,33 @@
 {
-  "axquant_version": "1.0.1",
   "calibration": null,
-  "created_at": "2026-08-02T19:52:56.105649Z",
-  "effective_bpw": 5.279832291155292,
   "files": [
     {
       "path": "README.md",
-      "sha256": "efee8e591496ed537a3fc7d0e8d2090a6fbefd03d670ef7fbaacb087b34befff",
-      "size_bytes": 7456
     },
     {
       "path": "axquant_plan.json",
-      "sha256": "02b20463cc90ca81dcd03492b246ab9b7f4c97bed3f248052deee83105b74144",
-      "size_bytes": 347258
     },
     {
       "path": "axquant_quantizer_execution.json",
-      "sha256": "c97f9283ba562ba735d4cefc191e35f7b7964318b9325c08d5ebd47fd0e2a76e",
       "size_bytes": 72744
     },
     {
       "path": "axquant_runtime.json",
-      "sha256": "5715430cce19e6ec6627c9d8e4d07c96cb935cf21582561db7734b22480f6042",
-      "size_bytes": 1643
     },
     {
       "path": "chat_template.jinja",
@@ -31,8 +36,8 @@
     },
     {
       "path": "config.json",
-      "sha256": "d444f78fddd3e95b0f2a18725a0ac72a923b79deb3adbe9f75b9b2abd1a92342",
-      "size_bytes": 87042
     },
     {
       "path": "generation_config.json",
@@ -41,36 +46,41 @@
     },
     {
       "path": "model-00001-of-00002.safetensors",
-      "sha256": "2581a3f8a5beb43f361c0dd2886cdff436bdbad90a2b46ca3e13864e3b1e1f41",
       "size_bytes": 5361479864
     },
     {
       "path": "model-00002-of-00002.safetensors",
-      "sha256": "7e6eba9b62bd4e3ed0035efc5254d961768e21915feb6c9adc84c52c6d91561d",
-      "size_bytes": 3552365268
     },
     {
       "path": "model.safetensors.index.json",
-      "sha256": "56d962482ac45a61674b07033f2a8b5a9ef1106568eddc5148dfcaa789132e9e",
       "size_bytes": 93118
     },
     {
       "path": "tokenizer.json",
-      "sha256": "99cf274236c60277fcfad861a5a1007518687ad06ba8938760f50b55ffa0b1ef",
-      "size_bytes": 17077420
     },
     {
       "path": "tokenizer_config.json",
-      "sha256": "790c3e17aefff6a4e0daca94a2dfbb17d4db65df0421ed3bc7759cda526b93c2",
-      "size_bytes": 21222
     }
   ],
   "format": "mlx",
-  "logical_parameters": 13506073600,
-  "main_logical_parameters": 13506073600,
-  "main_weight_file_size_bytes": 8913845132,
-  "measured_main_bpw": 5.2799031878517235,
-  "measured_total_bpw": 5.2799031878517235,
   "mtp_acceptance_retention": null,
   "mtp_distribution": {},
   "mtp_measured_speedup": null,
@@ -88,9 +98,9 @@
   },
   "mtp_present": false,
   "mtp_weight_file_size_bytes": 0,
-  "plan_sha256": "864e1704f89086e16e8e8b124ef72e348aea1b7ad863e8dd1a6af299af74e835",
   "profile": "general",
-  "protected_weight_file_size_bytes": 0,
   "quantizer": "axquant",
   "runtime": {
     "ax_engine": {
@@ -104,7 +114,7 @@
       {
         "compatibility_level": "B",
         "manifest": "config.json",
-        "mtp_support": "runtime-dependent",
         "name": "mlx-lm",
         "notes": [
           "Standard backbone inference is the compatibility target.",
@@ -115,7 +125,7 @@
         "support_level": "standard-inference"
       }
     ],
-    "created_at": "2026-08-02T19:52:51.472949Z",
     "kv_cache": null,
     "memory_policy": {
       "kv_cache_precision": "runtime-default",
@@ -139,7 +149,7 @@
     "primary_runtime": {
       "compatibility_level": "A",
       "manifest": "model-manifest.json",
-      "mtp_support": "native",
       "name": "ax-engine",
       "notes": [
         "Runtime claims require a passing AX Engine doctor and benchmark report."
@@ -153,7 +163,7 @@
   "schema_version": "axquant.artifact.v2",
   "software_versions": {
     "ax_engine": null,
-    "axquant": "1.0.1",
     "mlx": "0.32.0",
     "mlx_lm": "0.31.3",
     "pydantic": "2.13.4",
@@ -164,23 +174,23 @@
     "architecture": "Mistral3ForConditionalGeneration",
     "format": "mlx",
     "local_path": null,
-    "model_id": "mistralai/Ministral-3-14B-Instruct-2512",
-    "revision": null
   },
-  "target_class": "4bit",
   "weight_distribution": {
     "4bit": {
-      "fraction": 0.9005934633733967,
       "parameters": 12163481600
     },
     "8bit": {
-      "fraction": 0.049687915220601196,
       "parameters": 671088640
     },
     "bf16": {
-      "fraction": 0.04971862140600211,
-      "parameters": 671503360
     }
   },
-  "weight_file_size_bytes": 8913845132
 }

 {
+  "axquant_version": "1.2.0",
   "calibration": null,
+  "created_at": "2026-08-05T08:00:33.026860Z",
+  "effective_bpw": 5.6099475738157665,
   "files": [
     {
       "path": "README.md",
+      "sha256": "a5d1b0bda06d65e19f39437a968aca94f352d5a96bbbeccd040028f508dca66b",
+      "size_bytes": 8160
     },
     {
       "path": "axquant_plan.json",
+      "sha256": "68ff3d56821c0889b33c00297e0294172d93a59b8b69c1682e2c15eba94118b9",
+      "size_bytes": 544082
     },
     {
       "path": "axquant_quantizer_execution.json",
+      "sha256": "da8577ee59f6ac1ceef0b02b00a71d598baa6d607bd2af7b24d425c2c8fa0889",
       "size_bytes": 72744
     },
     {
       "path": "axquant_runtime.json",
+      "sha256": "3de6aeff0857f7bd58c76faaa75e20abfe32e3e8a3e13b746faaffd1fa05c9ae",
+      "size_bytes": 1628
+    },
+    {
+      "path": "axquant_vision_sidecar_manifest.json",
+      "sha256": "6e85bdfca944a1c61568958002d47f0c1c8e9a244af3c5e1b6642fb865b8af72",
+      "size_bytes": 916
     },
     {
       "path": "chat_template.jinja",
     },
     {
       "path": "config.json",
+      "sha256": "06e149fad49a02a9763f64902f4d06378ba52f59650dc2a7cff7b94b662b525c",
+      "size_bytes": 72666
     },
     {
       "path": "generation_config.json",
     },
     {
       "path": "model-00001-of-00002.safetensors",
+      "sha256": "f11db713f005b3eee6389231907ec6aa4679de7852df665e926b7178f96a9f86",
       "size_bytes": 5361479864
     },
     {
       "path": "model-00002-of-00002.safetensors",
+      "sha256": "b53c1dd8a5203e504fb3bc7f71adca58727adc7340fccbdb8a78ad312cb170f5",
+      "size_bytes": 3539585712
     },
     {
       "path": "model.safetensors.index.json",
+      "sha256": "9dede08d899bc5c3baf5b55bf7f68999d8ec5c3c687bc17936b9a7fd96387308",
       "size_bytes": 93118
     },
     {
       "path": "tokenizer.json",
+      "sha256": "d5f6046775b112f0e2d456ee9dba450684ab964fe5c4e231599bdc6773028135",
+      "size_bytes": 17078128
     },
     {
       "path": "tokenizer_config.json",
+      "sha256": "39f05766964b37275aa5e64fc812f3dbc7610cc6b2653fd5b6f7cf565ad94284",
+      "size_bytes": 468
+    },
+    {
+      "path": "vision.safetensors",
+      "sha256": "b1ae78df075fefa8f9e6eb5f157640530efd1bc51c5d07239145fa61a67d09ad",
+      "size_bytes": 877945912
     }
   ],
   "format": "mlx",
+  "logical_parameters": 13945031680,
+  "main_logical_parameters": 13945031680,
+  "main_weight_file_size_bytes": 9779011488,
+  "measured_main_bpw": 5.610033286349623,
+  "measured_total_bpw": 5.610033286349623,
   "mtp_acceptance_retention": null,
   "mtp_distribution": {},
   "mtp_measured_speedup": null,
   },
   "mtp_present": false,
   "mtp_weight_file_size_bytes": 0,
+  "plan_sha256": "b16e4310e1401539e785d2b2ffaab5bb0427ec7ab693cd96ff552b0a7ce6ef87",
   "profile": "general",
+  "protected_weight_file_size_bytes": 877945912,
   "quantizer": "axquant",
   "runtime": {
     "ax_engine": {
       {
         "compatibility_level": "B",
         "manifest": "config.json",
+        "mtp_support": "none",
         "name": "mlx-lm",
         "notes": [
           "Standard backbone inference is the compatibility target.",
         "support_level": "standard-inference"
       }
     ],
+    "created_at": "2026-08-05T08:00:22.646438Z",
     "kv_cache": null,
     "memory_policy": {
       "kv_cache_precision": "runtime-default",
     "primary_runtime": {
       "compatibility_level": "A",
       "manifest": "model-manifest.json",
+      "mtp_support": "none",
       "name": "ax-engine",
       "notes": [
         "Runtime claims require a passing AX Engine doctor and benchmark report."
   "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",
     "architecture": "Mistral3ForConditionalGeneration",
     "format": "mlx",
     "local_path": null,
+    "model_id": "mistralai/Ministral-3-14B-Instruct-2512-BF16",
+    "revision": "3cea74c1ebaf5ce5f5a2553de470e2ceab825142"
   },
+  "target_class": "5p6bpw",
   "weight_distribution": {
     "4bit": {
+      "fraction": 0.8722448165854579,
       "parameters": 12163481600
     },
     "8bit": {
+      "fraction": 0.0481238519495425,
       "parameters": 671088640
     },
     "bf16": {
+      "fraction": 0.07963133146499958,
+      "parameters": 1110461440
     }
   },
+  "weight_file_size_bytes": 9779011488
 }

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": "2026-08-02T19:52:51.464543Z",
-  "plan_sha256": "864e1704f89086e16e8e8b124ef72e348aea1b7ad863e8dd1a6af299af74e835",
   "records": [
     {
       "bits": 8,
@@ -298,7 +298,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -308,7 +308,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -318,7 +318,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -328,7 +328,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -338,7 +338,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -348,7 +348,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -358,7 +358,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.4.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -368,7 +368,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -378,7 +378,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -388,7 +388,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -398,7 +398,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -408,7 +408,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -418,7 +418,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -428,7 +428,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.5.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -438,7 +438,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -448,7 +448,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -458,7 +458,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -468,7 +468,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -478,7 +478,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -488,7 +488,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -498,7 +498,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -508,7 +508,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -518,7 +518,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -528,7 +528,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -538,7 +538,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -548,7 +548,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.10.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -558,7 +558,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -568,7 +568,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -578,7 +578,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -588,7 +588,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -598,7 +598,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -608,7 +608,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -618,7 +618,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.11.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -628,7 +628,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -638,7 +638,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -648,7 +648,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -658,7 +658,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -668,7 +668,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -678,7 +678,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -688,7 +688,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.12.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -698,7 +698,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -708,7 +708,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -718,7 +718,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -728,7 +728,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -738,7 +738,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -748,7 +748,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -758,7 +758,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.13.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -768,7 +768,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -778,7 +778,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -788,7 +788,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -798,7 +798,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -808,7 +808,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -818,7 +818,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -828,7 +828,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.14.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -838,7 +838,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -848,7 +848,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -858,7 +858,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -868,7 +868,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -878,7 +878,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -888,7 +888,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.6.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -898,7 +898,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -908,7 +908,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -918,7 +918,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -928,7 +928,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -938,7 +938,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -948,7 +948,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -958,7 +958,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.7.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -968,7 +968,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -978,7 +978,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -988,7 +988,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -998,7 +998,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1008,7 +1008,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1018,7 +1018,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1028,7 +1028,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.8.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1038,7 +1038,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1048,7 +1048,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1058,7 +1058,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1068,7 +1068,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1078,7 +1078,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1088,7 +1088,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1098,7 +1098,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.9.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1108,7 +1108,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1118,7 +1118,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1128,7 +1128,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.15.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1418,7 +1418,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1428,7 +1428,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1438,7 +1438,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1448,7 +1448,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1458,7 +1458,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1468,7 +1468,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1478,7 +1478,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.20.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1648,7 +1648,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.23.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1658,7 +1658,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.23.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1668,7 +1668,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.23.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1678,7 +1678,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.23.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1688,7 +1688,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.23.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1978,7 +1978,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1988,7 +1988,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -1998,7 +1998,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2008,7 +2008,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2018,7 +2018,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2028,7 +2028,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2038,7 +2038,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.28.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2258,7 +2258,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2268,7 +2268,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2278,7 +2278,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2288,7 +2288,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2298,7 +2298,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2308,7 +2308,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2318,7 +2318,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.32.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2425,7 +2425,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.34.self_attn.k_proj",
@@ -2455,7 +2455,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.34.self_attn.v_proj",
@@ -2495,7 +2495,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.k_proj",
@@ -2515,7 +2515,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.q_proj",
@@ -2525,7 +2525,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.v_proj",
@@ -2538,7 +2538,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2548,7 +2548,7 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2558,47 +2558,47 @@
       "group_size": 64,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
-      "module_path": "language_model.model.layers.36.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
@@ -2635,7 +2635,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.k_proj",
@@ -2645,7 +2645,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.o_proj",
@@ -2655,7 +2655,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.q_proj",
@@ -2665,7 +2665,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.v_proj",
@@ -2705,7 +2705,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.k_proj",
@@ -2715,7 +2715,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.o_proj",
@@ -2725,7 +2725,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.q_proj",
@@ -2735,7 +2735,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.v_proj",
@@ -2775,7 +2775,7 @@
     {
       "bits": 4,
       "fallback": false,
-      "group_size": 32,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.39.self_attn.k_proj",

 {
+  "created_at": "2026-08-05T08:00:18.943902Z",
+  "plan_sha256": "b16e4310e1401539e785d2b2ffaab5bb0427ec7ab693cd96ff552b0a7ce6ef87",
   "records": [
     {
       "bits": 8,
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.10.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.11.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.4.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.5.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.6.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.7.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.8.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.9.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.12.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.13.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.14.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.15.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.20.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.23.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.23.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.23.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.23.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.23.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.28.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.32.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.34.self_attn.k_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.34.self_attn.v_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.k_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.q_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.35.self_attn.v_proj",
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.mlp.gate_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.self_attn.k_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
       "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.self_attn.o_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.self_attn.q_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.self_attn.v_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.mlp.down_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
+      "module_path": "language_model.model.layers.36.mlp.up_proj",
       "note": "MLX-LM affine packing",
       "success": true
     },
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.k_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.o_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.q_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.37.self_attn.v_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.k_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.o_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.q_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.38.self_attn.v_proj",
     {
       "bits": 4,
       "fallback": false,
+      "group_size": 64,
       "metadata": {},
       "method": "affine",
       "module_path": "language_model.model.layers.39.self_attn.k_proj",

axquant_runtime.json CHANGED Viewed

@@ -10,7 +10,7 @@
     {
       "compatibility_level": "B",
       "manifest": "config.json",
-      "mtp_support": "runtime-dependent",
       "name": "mlx-lm",
       "notes": [
         "Standard backbone inference is the compatibility target.",
@@ -21,7 +21,7 @@
       "support_level": "standard-inference"
     }
   ],
-  "created_at": "2026-08-02T19:52:51.472949Z",
   "kv_cache": null,
   "memory_policy": {
     "kv_cache_precision": "runtime-default",
@@ -45,7 +45,7 @@
   "primary_runtime": {
     "compatibility_level": "A",
     "manifest": "model-manifest.json",
-    "mtp_support": "native",
     "name": "ax-engine",
     "notes": [
       "Runtime claims require a passing AX Engine doctor and benchmark report."

     {
       "compatibility_level": "B",
       "manifest": "config.json",
+      "mtp_support": "none",
       "name": "mlx-lm",
       "notes": [
         "Standard backbone inference is the compatibility target.",
       "support_level": "standard-inference"
     }
   ],
+  "created_at": "2026-08-05T08:00:22.646438Z",
   "kv_cache": null,
   "memory_policy": {
     "kv_cache_precision": "runtime-default",
   "primary_runtime": {
     "compatibility_level": "A",
     "manifest": "model-manifest.json",
+    "mtp_support": "none",
     "name": "ax-engine",
     "notes": [
       "Runtime claims require a passing AX Engine doctor and benchmark report."

axquant_vision_sidecar_manifest.json ADDED Viewed

	@@ -0,0 +1,30 @@

+{
+  "created_at": "2026-08-05T08:00:22.640676Z",
+  "dtypes": [
+    "BF16"
+  ],
+  "output": {
+    "path": "vision.safetensors",
+    "sha256": "b1ae78df075fefa8f9e6eb5f157640530efd1bc51c5d07239145fa61a67d09ad",
+    "size_bytes": 877945912
+  },
+  "parameters": 438958080,
+  "role": "vision",
+  "schema_version": "axquant.protected-tensor-sidecar.v1",
+  "source_files": [
+    {
+      "path": "model-00001-of-00006.safetensors",
+      "sha256": "a381935aa826a6960e752ab889f51b26870fa4ae34580dc5d33bba3b616a1e3b",
+      "size_bytes": 4925537056
+    }
+  ],
+  "source_model": {
+    "architecture": "Mistral3ForConditionalGeneration",
+    "format": "mlx",
+    "local_path": null,
+    "model_id": "mistralai/Ministral-3-14B-Instruct-2512-BF16",
+    "revision": "3cea74c1ebaf5ce5f5a2553de470e2ceab825142"
+  },
+  "tensor_count": 222,
+  "tensor_names_sha256": "3fa97c3c5b5b901ab47f5427584b61d7181707f7478b7eb50cec9f2aeeb04e7f"
+}

config.json CHANGED Viewed

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

model-00001-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2581a3f8a5beb43f361c0dd2886cdff436bdbad90a2b46ca3e13864e3b1e1f41
 size 5361479864

 version https://git-lfs.github.com/spec/v1
+oid sha256:f11db713f005b3eee6389231907ec6aa4679de7852df665e926b7178f96a9f86
 size 5361479864

model-00002-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e6eba9b62bd4e3ed0035efc5254d961768e21915feb6c9adc84c52c6d91561d
-size 3552365268

 version https://git-lfs.github.com/spec/v1
+oid sha256:b53c1dd8a5203e504fb3bc7f71adca58727adc7340fccbdb8a78ad312cb170f5
+size 3539585712

model.safetensors.index.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
     "metadata": {
-        "total_size": 8913725440,
         "total_parameters": 13506073600
     },
     "weight_map": {

 {
     "metadata": {
+        "total_size": 8900945920,
         "total_parameters": 13506073600
     },
     "weight_map": {

tokenizer.json CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99cf274236c60277fcfad861a5a1007518687ad06ba8938760f50b55ffa0b1ef
-size 17077420

 version https://git-lfs.github.com/spec/v1
+oid sha256:d5f6046775b112f0e2d456ee9dba450684ab964fe5c4e231599bdc6773028135
+size 17078128

tokenizer_config.json CHANGED Viewed

@@ -1,1015 +1,17 @@
 {
   "backend": "tokenizers",
   "bos_token": "<s>",
   "eos_token": "</s>",
-  "extra_special_tokens": [
-    "<unk>",
-    "<s>",
-    "</s>",
-    "[INST]",
-    "[/INST]",
-    "[AVAILABLE_TOOLS]",
-    "[/AVAILABLE_TOOLS]",
-    "[TOOL_RESULTS]",
-    "[/TOOL_RESULTS]",
-    "[TOOL_CALLS]",
-    "[IMG]",
-    "<pad>",
-    "[IMG_BREAK]",
-    "[IMG_END]",
-    "[PREFIX]",
-    "[MIDDLE]",
-    "[SUFFIX]",
-    "[SYSTEM_PROMPT]",
-    "[/SYSTEM_PROMPT]",
-    "[TOOL_CONTENT]",
-    "<SPECIAL_20>",
-    "<SPECIAL_21>",
-    "<SPECIAL_22>",
-    "<SPECIAL_23>",
-    "[AUDIO]",
-    "[BEGIN_AUDIO]",
-    "<SPECIAL_26>",
-    "<SPECIAL_27>",
-    "<SPECIAL_28>",
-    "<SPECIAL_29>",
-    "<SPECIAL_30>",
-    "<SPECIAL_31>",
-    "[ARGS]",
-    "[CALL_ID]",
-    "<SPECIAL_34>",
-    "<SPECIAL_35>",
-    "<SPECIAL_36>",
-    "<SPECIAL_37>",
-    "<SPECIAL_38>",
-    "<SPECIAL_39>",
-    "<SPECIAL_40>",
-    "<SPECIAL_41>",
-    "<SPECIAL_42>",
-    "<SPECIAL_43>",
-    "<SPECIAL_44>",
-    "<SPECIAL_45>",
-    "<SPECIAL_46>",
-    "<SPECIAL_47>",
-    "<SPECIAL_48>",
-    "<SPECIAL_49>",
-    "<SPECIAL_50>",
-    "<SPECIAL_51>",
-    "<SPECIAL_52>",
-    "<SPECIAL_53>",
-    "<SPECIAL_54>",
-    "<SPECIAL_55>",
-    "<SPECIAL_56>",
-    "<SPECIAL_57>",
-    "<SPECIAL_58>",
-    "<SPECIAL_59>",
-    "<SPECIAL_60>",
-    "<SPECIAL_61>",
-    "<SPECIAL_62>",
-    "<SPECIAL_63>",
-    "<SPECIAL_64>",
-    "<SPECIAL_65>",
-    "<SPECIAL_66>",
-    "<SPECIAL_67>",
-    "<SPECIAL_68>",
-    "<SPECIAL_69>",
-    "<SPECIAL_70>",
-    "<SPECIAL_71>",
-    "<SPECIAL_72>",
-    "<SPECIAL_73>",
-    "<SPECIAL_74>",
-    "<SPECIAL_75>",
-    "<SPECIAL_76>",
-    "<SPECIAL_77>",
-    "<SPECIAL_78>",
-    "<SPECIAL_79>",
-    "<SPECIAL_80>",
-    "<SPECIAL_81>",
-    "<SPECIAL_82>",
-    "<SPECIAL_83>",
-    "<SPECIAL_84>",
-    "<SPECIAL_85>",
-    "<SPECIAL_86>",
-    "<SPECIAL_87>",
-    "<SPECIAL_88>",
-    "<SPECIAL_89>",
-    "<SPECIAL_90>",
-    "<SPECIAL_91>",
-    "<SPECIAL_92>",
-    "<SPECIAL_93>",
-    "<SPECIAL_94>",
-    "<SPECIAL_95>",
-    "<SPECIAL_96>",
-    "<SPECIAL_97>",
-    "<SPECIAL_98>",
-    "<SPECIAL_99>",
-    "<SPECIAL_100>",
-    "<SPECIAL_101>",
-    "<SPECIAL_102>",
-    "<SPECIAL_103>",
-    "<SPECIAL_104>",
-    "<SPECIAL_105>",
-    "<SPECIAL_106>",
-    "<SPECIAL_107>",
-    "<SPECIAL_108>",
-    "<SPECIAL_109>",
-    "<SPECIAL_110>",
-    "<SPECIAL_111>",
-    "<SPECIAL_112>",
-    "<SPECIAL_113>",
-    "<SPECIAL_114>",
-    "<SPECIAL_115>",
-    "<SPECIAL_116>",
-    "<SPECIAL_117>",
-    "<SPECIAL_118>",
-    "<SPECIAL_119>",
-    "<SPECIAL_120>",
-    "<SPECIAL_121>",
-    "<SPECIAL_122>",
-    "<SPECIAL_123>",
-    "<SPECIAL_124>",
-    "<SPECIAL_125>",
-    "<SPECIAL_126>",
-    "<SPECIAL_127>",
-    "<SPECIAL_128>",
-    "<SPECIAL_129>",
-    "<SPECIAL_130>",
-    "<SPECIAL_131>",
-    "<SPECIAL_132>",
-    "<SPECIAL_133>",
-    "<SPECIAL_134>",
-    "<SPECIAL_135>",
-    "<SPECIAL_136>",
-    "<SPECIAL_137>",
-    "<SPECIAL_138>",
-    "<SPECIAL_139>",
-    "<SPECIAL_140>",
-    "<SPECIAL_141>",
-    "<SPECIAL_142>",
-    "<SPECIAL_143>",
-    "<SPECIAL_144>",
-    "<SPECIAL_145>",
-    "<SPECIAL_146>",
-    "<SPECIAL_147>",
-    "<SPECIAL_148>",
-    "<SPECIAL_149>",
-    "<SPECIAL_150>",
-    "<SPECIAL_151>",
-    "<SPECIAL_152>",
-    "<SPECIAL_153>",
-    "<SPECIAL_154>",
-    "<SPECIAL_155>",
-    "<SPECIAL_156>",
-    "<SPECIAL_157>",
-    "<SPECIAL_158>",
-    "<SPECIAL_159>",
-    "<SPECIAL_160>",
-    "<SPECIAL_161>",
-    "<SPECIAL_162>",
-    "<SPECIAL_163>",
-    "<SPECIAL_164>",
-    "<SPECIAL_165>",
-    "<SPECIAL_166>",
-    "<SPECIAL_167>",
-    "<SPECIAL_168>",
-    "<SPECIAL_169>",
-    "<SPECIAL_170>",
-    "<SPECIAL_171>",
-    "<SPECIAL_172>",
-    "<SPECIAL_173>",
-    "<SPECIAL_174>",
-    "<SPECIAL_175>",
-    "<SPECIAL_176>",
-    "<SPECIAL_177>",
-    "<SPECIAL_178>",
-    "<SPECIAL_179>",
-    "<SPECIAL_180>",
-    "<SPECIAL_181>",
-    "<SPECIAL_182>",
-    "<SPECIAL_183>",
-    "<SPECIAL_184>",
-    "<SPECIAL_185>",
-    "<SPECIAL_186>",
-    "<SPECIAL_187>",
-    "<SPECIAL_188>",
-    "<SPECIAL_189>",
-    "<SPECIAL_190>",
-    "<SPECIAL_191>",
-    "<SPECIAL_192>",
-    "<SPECIAL_193>",
-    "<SPECIAL_194>",
-    "<SPECIAL_195>",
-    "<SPECIAL_196>",
-    "<SPECIAL_197>",
-    "<SPECIAL_198>",
-    "<SPECIAL_199>",
-    "<SPECIAL_200>",
-    "<SPECIAL_201>",
-    "<SPECIAL_202>",
-    "<SPECIAL_203>",
-    "<SPECIAL_204>",
-    "<SPECIAL_205>",
-    "<SPECIAL_206>",
-    "<SPECIAL_207>",
-    "<SPECIAL_208>",
-    "<SPECIAL_209>",
-    "<SPECIAL_210>",
-    "<SPECIAL_211>",
-    "<SPECIAL_212>",
-    "<SPECIAL_213>",
-    "<SPECIAL_214>",
-    "<SPECIAL_215>",
-    "<SPECIAL_216>",
-    "<SPECIAL_217>",
-    "<SPECIAL_218>",
-    "<SPECIAL_219>",
-    "<SPECIAL_220>",
-    "<SPECIAL_221>",
-    "<SPECIAL_222>",
-    "<SPECIAL_223>",
-    "<SPECIAL_224>",
-    "<SPECIAL_225>",
-    "<SPECIAL_226>",
-    "<SPECIAL_227>",
-    "<SPECIAL_228>",
-    "<SPECIAL_229>",
-    "<SPECIAL_230>",
-    "<SPECIAL_231>",
-    "<SPECIAL_232>",
-    "<SPECIAL_233>",
-    "<SPECIAL_234>",
-    "<SPECIAL_235>",
-    "<SPECIAL_236>",
-    "<SPECIAL_237>",
-    "<SPECIAL_238>",
-    "<SPECIAL_239>",
-    "<SPECIAL_240>",
-    "<SPECIAL_241>",
-    "<SPECIAL_242>",
-    "<SPECIAL_243>",
-    "<SPECIAL_244>",
-    "<SPECIAL_245>",
-    "<SPECIAL_246>",
-    "<SPECIAL_247>",
-    "<SPECIAL_248>",
-    "<SPECIAL_249>",
-    "<SPECIAL_250>",
-    "<SPECIAL_251>",
-    "<SPECIAL_252>",
-    "<SPECIAL_253>",
-    "<SPECIAL_254>",
-    "<SPECIAL_255>",
-    "<SPECIAL_256>",
-    "<SPECIAL_257>",
-    "<SPECIAL_258>",
-    "<SPECIAL_259>",
-    "<SPECIAL_260>",
-    "<SPECIAL_261>",
-    "<SPECIAL_262>",
-    "<SPECIAL_263>",
-    "<SPECIAL_264>",
-    "<SPECIAL_265>",
-    "<SPECIAL_266>",
-    "<SPECIAL_267>",
-    "<SPECIAL_268>",
-    "<SPECIAL_269>",
-    "<SPECIAL_270>",
-    "<SPECIAL_271>",
-    "<SPECIAL_272>",
-    "<SPECIAL_273>",
-    "<SPECIAL_274>",
-    "<SPECIAL_275>",
-    "<SPECIAL_276>",
-    "<SPECIAL_277>",
-    "<SPECIAL_278>",
-    "<SPECIAL_279>",
-    "<SPECIAL_280>",
-    "<SPECIAL_281>",
-    "<SPECIAL_282>",
-    "<SPECIAL_283>",
-    "<SPECIAL_284>",
-    "<SPECIAL_285>",
-    "<SPECIAL_286>",
-    "<SPECIAL_287>",
-    "<SPECIAL_288>",
-    "<SPECIAL_289>",
-    "<SPECIAL_290>",
-    "<SPECIAL_291>",
-    "<SPECIAL_292>",
-    "<SPECIAL_293>",
-    "<SPECIAL_294>",
-    "<SPECIAL_295>",
-    "<SPECIAL_296>",
-    "<SPECIAL_297>",
-    "<SPECIAL_298>",
-    "<SPECIAL_299>",
-    "<SPECIAL_300>",
-    "<SPECIAL_301>",
-    "<SPECIAL_302>",
-    "<SPECIAL_303>",
-    "<SPECIAL_304>",
-    "<SPECIAL_305>",
-    "<SPECIAL_306>",
-    "<SPECIAL_307>",
-    "<SPECIAL_308>",
-    "<SPECIAL_309>",
-    "<SPECIAL_310>",
-    "<SPECIAL_311>",
-    "<SPECIAL_312>",
-    "<SPECIAL_313>",
-    "<SPECIAL_314>",
-    "<SPECIAL_315>",
-    "<SPECIAL_316>",
-    "<SPECIAL_317>",
-    "<SPECIAL_318>",
-    "<SPECIAL_319>",
-    "<SPECIAL_320>",
-    "<SPECIAL_321>",
-    "<SPECIAL_322>",
-    "<SPECIAL_323>",
-    "<SPECIAL_324>",
-    "<SPECIAL_325>",
-    "<SPECIAL_326>",
-    "<SPECIAL_327>",
-    "<SPECIAL_328>",
-    "<SPECIAL_329>",
-    "<SPECIAL_330>",
-    "<SPECIAL_331>",
-    "<SPECIAL_332>",
-    "<SPECIAL_333>",
-    "<SPECIAL_334>",
-    "<SPECIAL_335>",
-    "<SPECIAL_336>",
-    "<SPECIAL_337>",
-    "<SPECIAL_338>",
-    "<SPECIAL_339>",
-    "<SPECIAL_340>",
-    "<SPECIAL_341>",
-    "<SPECIAL_342>",
-    "<SPECIAL_343>",
-    "<SPECIAL_344>",
-    "<SPECIAL_345>",
-    "<SPECIAL_346>",
-    "<SPECIAL_347>",
-    "<SPECIAL_348>",
-    "<SPECIAL_349>",
-    "<SPECIAL_350>",
-    "<SPECIAL_351>",
-    "<SPECIAL_352>",
-    "<SPECIAL_353>",
-    "<SPECIAL_354>",
-    "<SPECIAL_355>",
-    "<SPECIAL_356>",
-    "<SPECIAL_357>",
-    "<SPECIAL_358>",
-    "<SPECIAL_359>",
-    "<SPECIAL_360>",
-    "<SPECIAL_361>",
-    "<SPECIAL_362>",
-    "<SPECIAL_363>",
-    "<SPECIAL_364>",
-    "<SPECIAL_365>",
-    "<SPECIAL_366>",
-    "<SPECIAL_367>",
-    "<SPECIAL_368>",
-    "<SPECIAL_369>",
-    "<SPECIAL_370>",
-    "<SPECIAL_371>",
-    "<SPECIAL_372>",
-    "<SPECIAL_373>",
-    "<SPECIAL_374>",
-    "<SPECIAL_375>",
-    "<SPECIAL_376>",
-    "<SPECIAL_377>",
-    "<SPECIAL_378>",
-    "<SPECIAL_379>",
-    "<SPECIAL_380>",
-    "<SPECIAL_381>",
-    "<SPECIAL_382>",
-    "<SPECIAL_383>",
-    "<SPECIAL_384>",
-    "<SPECIAL_385>",
-    "<SPECIAL_386>",
-    "<SPECIAL_387>",
-    "<SPECIAL_388>",
-    "<SPECIAL_389>",
-    "<SPECIAL_390>",
-    "<SPECIAL_391>",
-    "<SPECIAL_392>",
-    "<SPECIAL_393>",
-    "<SPECIAL_394>",
-    "<SPECIAL_395>",
-    "<SPECIAL_396>",
-    "<SPECIAL_397>",
-    "<SPECIAL_398>",
-    "<SPECIAL_399>",
-    "<SPECIAL_400>",
-    "<SPECIAL_401>",
-    "<SPECIAL_402>",
-    "<SPECIAL_403>",
-    "<SPECIAL_404>",
-    "<SPECIAL_405>",
-    "<SPECIAL_406>",
-    "<SPECIAL_407>",
-    "<SPECIAL_408>",
-    "<SPECIAL_409>",
-    "<SPECIAL_410>",
-    "<SPECIAL_411>",
-    "<SPECIAL_412>",
-    "<SPECIAL_413>",
-    "<SPECIAL_414>",
-    "<SPECIAL_415>",
-    "<SPECIAL_416>",
-    "<SPECIAL_417>",
-    "<SPECIAL_418>",
-    "<SPECIAL_419>",
-    "<SPECIAL_420>",
-    "<SPECIAL_421>",
-    "<SPECIAL_422>",
-    "<SPECIAL_423>",
-    "<SPECIAL_424>",
-    "<SPECIAL_425>",
-    "<SPECIAL_426>",
-    "<SPECIAL_427>",
-    "<SPECIAL_428>",
-    "<SPECIAL_429>",
-    "<SPECIAL_430>",
-    "<SPECIAL_431>",
-    "<SPECIAL_432>",
-    "<SPECIAL_433>",
-    "<SPECIAL_434>",
-    "<SPECIAL_435>",
-    "<SPECIAL_436>",
-    "<SPECIAL_437>",
-    "<SPECIAL_438>",
-    "<SPECIAL_439>",
-    "<SPECIAL_440>",
-    "<SPECIAL_441>",
-    "<SPECIAL_442>",
-    "<SPECIAL_443>",
-    "<SPECIAL_444>",
-    "<SPECIAL_445>",
-    "<SPECIAL_446>",
-    "<SPECIAL_447>",
-    "<SPECIAL_448>",
-    "<SPECIAL_449>",
-    "<SPECIAL_450>",
-    "<SPECIAL_451>",
-    "<SPECIAL_452>",
-    "<SPECIAL_453>",
-    "<SPECIAL_454>",
-    "<SPECIAL_455>",
-    "<SPECIAL_456>",
-    "<SPECIAL_457>",
-    "<SPECIAL_458>",
-    "<SPECIAL_459>",
-    "<SPECIAL_460>",
-    "<SPECIAL_461>",
-    "<SPECIAL_462>",
-    "<SPECIAL_463>",
-    "<SPECIAL_464>",
-    "<SPECIAL_465>",
-    "<SPECIAL_466>",
-    "<SPECIAL_467>",
-    "<SPECIAL_468>",
-    "<SPECIAL_469>",
-    "<SPECIAL_470>",
-    "<SPECIAL_471>",
-    "<SPECIAL_472>",
-    "<SPECIAL_473>",
-    "<SPECIAL_474>",
-    "<SPECIAL_475>",
-    "<SPECIAL_476>",
-    "<SPECIAL_477>",
-    "<SPECIAL_478>",
-    "<SPECIAL_479>",
-    "<SPECIAL_480>",
-    "<SPECIAL_481>",
-    "<SPECIAL_482>",
-    "<SPECIAL_483>",
-    "<SPECIAL_484>",
-    "<SPECIAL_485>",
-    "<SPECIAL_486>",
-    "<SPECIAL_487>",
-    "<SPECIAL_488>",
-    "<SPECIAL_489>",
-    "<SPECIAL_490>",
-    "<SPECIAL_491>",
-    "<SPECIAL_492>",
-    "<SPECIAL_493>",
-    "<SPECIAL_494>",
-    "<SPECIAL_495>",
-    "<SPECIAL_496>",
-    "<SPECIAL_497>",
-    "<SPECIAL_498>",
-    "<SPECIAL_499>",
-    "<SPECIAL_500>",
-    "<SPECIAL_501>",
-    "<SPECIAL_502>",
-    "<SPECIAL_503>",
-    "<SPECIAL_504>",
-    "<SPECIAL_505>",
-    "<SPECIAL_506>",
-    "<SPECIAL_507>",
-    "<SPECIAL_508>",
-    "<SPECIAL_509>",
-    "<SPECIAL_510>",
-    "<SPECIAL_511>",
-    "<SPECIAL_512>",
-    "<SPECIAL_513>",
-    "<SPECIAL_514>",
-    "<SPECIAL_515>",
-    "<SPECIAL_516>",
-    "<SPECIAL_517>",
-    "<SPECIAL_518>",
-    "<SPECIAL_519>",
-    "<SPECIAL_520>",
-    "<SPECIAL_521>",
-    "<SPECIAL_522>",
-    "<SPECIAL_523>",
-    "<SPECIAL_524>",
-    "<SPECIAL_525>",
-    "<SPECIAL_526>",
-    "<SPECIAL_527>",
-    "<SPECIAL_528>",
-    "<SPECIAL_529>",
-    "<SPECIAL_530>",
-    "<SPECIAL_531>",
-    "<SPECIAL_532>",
-    "<SPECIAL_533>",
-    "<SPECIAL_534>",
-    "<SPECIAL_535>",
-    "<SPECIAL_536>",
-    "<SPECIAL_537>",
-    "<SPECIAL_538>",
-    "<SPECIAL_539>",
-    "<SPECIAL_540>",
-    "<SPECIAL_541>",
-    "<SPECIAL_542>",
-    "<SPECIAL_543>",
-    "<SPECIAL_544>",
-    "<SPECIAL_545>",
-    "<SPECIAL_546>",
-    "<SPECIAL_547>",
-    "<SPECIAL_548>",
-    "<SPECIAL_549>",
-    "<SPECIAL_550>",
-    "<SPECIAL_551>",
-    "<SPECIAL_552>",
-    "<SPECIAL_553>",
-    "<SPECIAL_554>",
-    "<SPECIAL_555>",
-    "<SPECIAL_556>",
-    "<SPECIAL_557>",
-    "<SPECIAL_558>",
-    "<SPECIAL_559>",
-    "<SPECIAL_560>",
-    "<SPECIAL_561>",
-    "<SPECIAL_562>",
-    "<SPECIAL_563>",
-    "<SPECIAL_564>",
-    "<SPECIAL_565>",
-    "<SPECIAL_566>",
-    "<SPECIAL_567>",
-    "<SPECIAL_568>",
-    "<SPECIAL_569>",
-    "<SPECIAL_570>",
-    "<SPECIAL_571>",
-    "<SPECIAL_572>",
-    "<SPECIAL_573>",
-    "<SPECIAL_574>",
-    "<SPECIAL_575>",
-    "<SPECIAL_576>",
-    "<SPECIAL_577>",
-    "<SPECIAL_578>",
-    "<SPECIAL_579>",
-    "<SPECIAL_580>",
-    "<SPECIAL_581>",
-    "<SPECIAL_582>",
-    "<SPECIAL_583>",
-    "<SPECIAL_584>",
-    "<SPECIAL_585>",
-    "<SPECIAL_586>",
-    "<SPECIAL_587>",
-    "<SPECIAL_588>",
-    "<SPECIAL_589>",
-    "<SPECIAL_590>",
-    "<SPECIAL_591>",
-    "<SPECIAL_592>",
-    "<SPECIAL_593>",
-    "<SPECIAL_594>",
-    "<SPECIAL_595>",
-    "<SPECIAL_596>",
-    "<SPECIAL_597>",
-    "<SPECIAL_598>",
-    "<SPECIAL_599>",
-    "<SPECIAL_600>",
-    "<SPECIAL_601>",
-    "<SPECIAL_602>",
-    "<SPECIAL_603>",
-    "<SPECIAL_604>",
-    "<SPECIAL_605>",
-    "<SPECIAL_606>",
-    "<SPECIAL_607>",
-    "<SPECIAL_608>",
-    "<SPECIAL_609>",
-    "<SPECIAL_610>",
-    "<SPECIAL_611>",
-    "<SPECIAL_612>",
-    "<SPECIAL_613>",
-    "<SPECIAL_614>",
-    "<SPECIAL_615>",
-    "<SPECIAL_616>",
-    "<SPECIAL_617>",
-    "<SPECIAL_618>",
-    "<SPECIAL_619>",
-    "<SPECIAL_620>",
-    "<SPECIAL_621>",
-    "<SPECIAL_622>",
-    "<SPECIAL_623>",
-    "<SPECIAL_624>",
-    "<SPECIAL_625>",
-    "<SPECIAL_626>",
-    "<SPECIAL_627>",
-    "<SPECIAL_628>",
-    "<SPECIAL_629>",
-    "<SPECIAL_630>",
-    "<SPECIAL_631>",
-    "<SPECIAL_632>",
-    "<SPECIAL_633>",
-    "<SPECIAL_634>",
-    "<SPECIAL_635>",
-    "<SPECIAL_636>",
-    "<SPECIAL_637>",
-    "<SPECIAL_638>",
-    "<SPECIAL_639>",
-    "<SPECIAL_640>",
-    "<SPECIAL_641>",
-    "<SPECIAL_642>",
-    "<SPECIAL_643>",
-    "<SPECIAL_644>",
-    "<SPECIAL_645>",
-    "<SPECIAL_646>",
-    "<SPECIAL_647>",
-    "<SPECIAL_648>",
-    "<SPECIAL_649>",
-    "<SPECIAL_650>",
-    "<SPECIAL_651>",
-    "<SPECIAL_652>",
-    "<SPECIAL_653>",
-    "<SPECIAL_654>",
-    "<SPECIAL_655>",
-    "<SPECIAL_656>",
-    "<SPECIAL_657>",
-    "<SPECIAL_658>",
-    "<SPECIAL_659>",
-    "<SPECIAL_660>",
-    "<SPECIAL_661>",
-    "<SPECIAL_662>",
-    "<SPECIAL_663>",
-    "<SPECIAL_664>",
-    "<SPECIAL_665>",
-    "<SPECIAL_666>",
-    "<SPECIAL_667>",
-    "<SPECIAL_668>",
-    "<SPECIAL_669>",
-    "<SPECIAL_670>",
-    "<SPECIAL_671>",
-    "<SPECIAL_672>",
-    "<SPECIAL_673>",
-    "<SPECIAL_674>",
-    "<SPECIAL_675>",
-    "<SPECIAL_676>",
-    "<SPECIAL_677>",
-    "<SPECIAL_678>",
-    "<SPECIAL_679>",
-    "<SPECIAL_680>",
-    "<SPECIAL_681>",
-    "<SPECIAL_682>",
-    "<SPECIAL_683>",
-    "<SPECIAL_684>",
-    "<SPECIAL_685>",
-    "<SPECIAL_686>",
-    "<SPECIAL_687>",
-    "<SPECIAL_688>",
-    "<SPECIAL_689>",
-    "<SPECIAL_690>",
-    "<SPECIAL_691>",
-    "<SPECIAL_692>",
-    "<SPECIAL_693>",
-    "<SPECIAL_694>",
-    "<SPECIAL_695>",
-    "<SPECIAL_696>",
-    "<SPECIAL_697>",
-    "<SPECIAL_698>",
-    "<SPECIAL_699>",
-    "<SPECIAL_700>",
-    "<SPECIAL_701>",
-    "<SPECIAL_702>",
-    "<SPECIAL_703>",
-    "<SPECIAL_704>",
-    "<SPECIAL_705>",
-    "<SPECIAL_706>",
-    "<SPECIAL_707>",
-    "<SPECIAL_708>",
-    "<SPECIAL_709>",
-    "<SPECIAL_710>",
-    "<SPECIAL_711>",
-    "<SPECIAL_712>",
-    "<SPECIAL_713>",
-    "<SPECIAL_714>",
-    "<SPECIAL_715>",
-    "<SPECIAL_716>",
-    "<SPECIAL_717>",
-    "<SPECIAL_718>",
-    "<SPECIAL_719>",
-    "<SPECIAL_720>",
-    "<SPECIAL_721>",
-    "<SPECIAL_722>",
-    "<SPECIAL_723>",
-    "<SPECIAL_724>",
-    "<SPECIAL_725>",
-    "<SPECIAL_726>",
-    "<SPECIAL_727>",
-    "<SPECIAL_728>",
-    "<SPECIAL_729>",
-    "<SPECIAL_730>",
-    "<SPECIAL_731>",
-    "<SPECIAL_732>",
-    "<SPECIAL_733>",
-    "<SPECIAL_734>",
-    "<SPECIAL_735>",
-    "<SPECIAL_736>",
-    "<SPECIAL_737>",
-    "<SPECIAL_738>",
-    "<SPECIAL_739>",
-    "<SPECIAL_740>",
-    "<SPECIAL_741>",
-    "<SPECIAL_742>",
-    "<SPECIAL_743>",
-    "<SPECIAL_744>",
-    "<SPECIAL_745>",
-    "<SPECIAL_746>",
-    "<SPECIAL_747>",
-    "<SPECIAL_748>",
-    "<SPECIAL_749>",
-    "<SPECIAL_750>",
-    "<SPECIAL_751>",
-    "<SPECIAL_752>",
-    "<SPECIAL_753>",
-    "<SPECIAL_754>",
-    "<SPECIAL_755>",
-    "<SPECIAL_756>",
-    "<SPECIAL_757>",
-    "<SPECIAL_758>",
-    "<SPECIAL_759>",
-    "<SPECIAL_760>",
-    "<SPECIAL_761>",
-    "<SPECIAL_762>",
-    "<SPECIAL_763>",
-    "<SPECIAL_764>",
-    "<SPECIAL_765>",
-    "<SPECIAL_766>",
-    "<SPECIAL_767>",
-    "<SPECIAL_768>",
-    "<SPECIAL_769>",
-    "<SPECIAL_770>",
-    "<SPECIAL_771>",
-    "<SPECIAL_772>",
-    "<SPECIAL_773>",
-    "<SPECIAL_774>",
-    "<SPECIAL_775>",
-    "<SPECIAL_776>",
-    "<SPECIAL_777>",
-    "<SPECIAL_778>",
-    "<SPECIAL_779>",
-    "<SPECIAL_780>",
-    "<SPECIAL_781>",
-    "<SPECIAL_782>",
-    "<SPECIAL_783>",
-    "<SPECIAL_784>",
-    "<SPECIAL_785>",
-    "<SPECIAL_786>",
-    "<SPECIAL_787>",
-    "<SPECIAL_788>",
-    "<SPECIAL_789>",
-    "<SPECIAL_790>",
-    "<SPECIAL_791>",
-    "<SPECIAL_792>",
-    "<SPECIAL_793>",
-    "<SPECIAL_794>",
-    "<SPECIAL_795>",
-    "<SPECIAL_796>",
-    "<SPECIAL_797>",
-    "<SPECIAL_798>",
-    "<SPECIAL_799>",
-    "<SPECIAL_800>",
-    "<SPECIAL_801>",
-    "<SPECIAL_802>",
-    "<SPECIAL_803>",
-    "<SPECIAL_804>",
-    "<SPECIAL_805>",
-    "<SPECIAL_806>",
-    "<SPECIAL_807>",
-    "<SPECIAL_808>",
-    "<SPECIAL_809>",
-    "<SPECIAL_810>",
-    "<SPECIAL_811>",
-    "<SPECIAL_812>",
-    "<SPECIAL_813>",
-    "<SPECIAL_814>",
-    "<SPECIAL_815>",
-    "<SPECIAL_816>",
-    "<SPECIAL_817>",
-    "<SPECIAL_818>",
-    "<SPECIAL_819>",
-    "<SPECIAL_820>",
-    "<SPECIAL_821>",
-    "<SPECIAL_822>",
-    "<SPECIAL_823>",
-    "<SPECIAL_824>",
-    "<SPECIAL_825>",
-    "<SPECIAL_826>",
-    "<SPECIAL_827>",
-    "<SPECIAL_828>",
-    "<SPECIAL_829>",
-    "<SPECIAL_830>",
-    "<SPECIAL_831>",
-    "<SPECIAL_832>",
-    "<SPECIAL_833>",
-    "<SPECIAL_834>",
-    "<SPECIAL_835>",
-    "<SPECIAL_836>",
-    "<SPECIAL_837>",
-    "<SPECIAL_838>",
-    "<SPECIAL_839>",
-    "<SPECIAL_840>",
-    "<SPECIAL_841>",
-    "<SPECIAL_842>",
-    "<SPECIAL_843>",
-    "<SPECIAL_844>",
-    "<SPECIAL_845>",
-    "<SPECIAL_846>",
-    "<SPECIAL_847>",
-    "<SPECIAL_848>",
-    "<SPECIAL_849>",
-    "<SPECIAL_850>",
-    "<SPECIAL_851>",
-    "<SPECIAL_852>",
-    "<SPECIAL_853>",
-    "<SPECIAL_854>",
-    "<SPECIAL_855>",
-    "<SPECIAL_856>",
-    "<SPECIAL_857>",
-    "<SPECIAL_858>",
-    "<SPECIAL_859>",
-    "<SPECIAL_860>",
-    "<SPECIAL_861>",
-    "<SPECIAL_862>",
-    "<SPECIAL_863>",
-    "<SPECIAL_864>",
-    "<SPECIAL_865>",
-    "<SPECIAL_866>",
-    "<SPECIAL_867>",
-    "<SPECIAL_868>",
-    "<SPECIAL_869>",
-    "<SPECIAL_870>",
-    "<SPECIAL_871>",
-    "<SPECIAL_872>",
-    "<SPECIAL_873>",
-    "<SPECIAL_874>",
-    "<SPECIAL_875>",
-    "<SPECIAL_876>",
-    "<SPECIAL_877>",
-    "<SPECIAL_878>",
-    "<SPECIAL_879>",
-    "<SPECIAL_880>",
-    "<SPECIAL_881>",
-    "<SPECIAL_882>",
-    "<SPECIAL_883>",
-    "<SPECIAL_884>",
-    "<SPECIAL_885>",
-    "<SPECIAL_886>",
-    "<SPECIAL_887>",
-    "<SPECIAL_888>",
-    "<SPECIAL_889>",
-    "<SPECIAL_890>",
-    "<SPECIAL_891>",
-    "<SPECIAL_892>",
-    "<SPECIAL_893>",
-    "<SPECIAL_894>",
-    "<SPECIAL_895>",
-    "<SPECIAL_896>",
-    "<SPECIAL_897>",
-    "<SPECIAL_898>",
-    "<SPECIAL_899>",
-    "<SPECIAL_900>",
-    "<SPECIAL_901>",
-    "<SPECIAL_902>",
-    "<SPECIAL_903>",
-    "<SPECIAL_904>",
-    "<SPECIAL_905>",
-    "<SPECIAL_906>",
-    "<SPECIAL_907>",
-    "<SPECIAL_908>",
-    "<SPECIAL_909>",
-    "<SPECIAL_910>",
-    "<SPECIAL_911>",
-    "<SPECIAL_912>",
-    "<SPECIAL_913>",
-    "<SPECIAL_914>",
-    "<SPECIAL_915>",
-    "<SPECIAL_916>",
-    "<SPECIAL_917>",
-    "<SPECIAL_918>",
-    "<SPECIAL_919>",
-    "<SPECIAL_920>",
-    "<SPECIAL_921>",
-    "<SPECIAL_922>",
-    "<SPECIAL_923>",
-    "<SPECIAL_924>",
-    "<SPECIAL_925>",
-    "<SPECIAL_926>",
-    "<SPECIAL_927>",
-    "<SPECIAL_928>",
-    "<SPECIAL_929>",
-    "<SPECIAL_930>",
-    "<SPECIAL_931>",
-    "<SPECIAL_932>",
-    "<SPECIAL_933>",
-    "<SPECIAL_934>",
-    "<SPECIAL_935>",
-    "<SPECIAL_936>",
-    "<SPECIAL_937>",
-    "<SPECIAL_938>",
-    "<SPECIAL_939>",
-    "<SPECIAL_940>",
-    "<SPECIAL_941>",
-    "<SPECIAL_942>",
-    "<SPECIAL_943>",
-    "<SPECIAL_944>",
-    "<SPECIAL_945>",
-    "<SPECIAL_946>",
-    "<SPECIAL_947>",
-    "<SPECIAL_948>",
-    "<SPECIAL_949>",
-    "<SPECIAL_950>",
-    "<SPECIAL_951>",
-    "<SPECIAL_952>",
-    "<SPECIAL_953>",
-    "<SPECIAL_954>",
-    "<SPECIAL_955>",
-    "<SPECIAL_956>",
-    "<SPECIAL_957>",
-    "<SPECIAL_958>",
-    "<SPECIAL_959>",
-    "<SPECIAL_960>",
-    "<SPECIAL_961>",
-    "<SPECIAL_962>",
-    "<SPECIAL_963>",
-    "<SPECIAL_964>",
-    "<SPECIAL_965>",
-    "<SPECIAL_966>",
-    "<SPECIAL_967>",
-    "<SPECIAL_968>",
-    "<SPECIAL_969>",
-    "<SPECIAL_970>",
-    "<SPECIAL_971>",
-    "<SPECIAL_972>",
-    "<SPECIAL_973>",
-    "<SPECIAL_974>",
-    "<SPECIAL_975>",
-    "<SPECIAL_976>",
-    "<SPECIAL_977>",
-    "<SPECIAL_978>",
-    "<SPECIAL_979>",
-    "<SPECIAL_980>",
-    "<SPECIAL_981>",
-    "<SPECIAL_982>",
-    "<SPECIAL_983>",
-    "<SPECIAL_984>",
-    "<SPECIAL_985>",
-    "<SPECIAL_986>",
-    "<SPECIAL_987>",
-    "<SPECIAL_988>",
-    "<SPECIAL_989>",
-    "<SPECIAL_990>",
-    "<SPECIAL_991>",
-    "<SPECIAL_992>",
-    "<SPECIAL_993>",
-    "<SPECIAL_994>",
-    "<SPECIAL_995>",
-    "<SPECIAL_996>",
-    "<SPECIAL_997>",
-    "<SPECIAL_998>",
-    "<SPECIAL_999>"
-  ],
   "is_local": true,
   "local_files_only": false,
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "processor_class": "PixtralProcessor",
   "tokenizer_class": "TokenizersBackend",
   "tool_parser_type": "mistral",
-  "unk_token": "<unk>"
 }

 {
+  "add_prefix_space": null,
   "backend": "tokenizers",
   "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "is_local": true,
+  "legacy": true,
   "local_files_only": false,
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<pad>",
   "processor_class": "PixtralProcessor",
   "tokenizer_class": "TokenizersBackend",
   "tool_parser_type": "mistral",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
 }

vision.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b1ae78df075fefa8f9e6eb5f157640530efd1bc51c5d07239145fa61a67d09ad
+size 877945912