Instructions to use jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k 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("jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k")

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 jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k 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 "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"

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 jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k"

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 "jedisct1/Qwen3.8-Flash-Next-oQ4e-MTP-128k" \
  --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"

jedisct1 commited on about 3 hours ago

Commit

9c6093c

verified ·

1 Parent(s): c0e307f

Correct model card details and indexed payload size

Browse files

Files changed (2) hide show

README.md +2 -2
model.safetensors.index.json +1 -1

README.md CHANGED Viewed

@@ -42,7 +42,7 @@ The model uses oMLX enhanced importance-matrix quantization with 4-bit affine, g
 - MTP attention, shared-expert, and layer mixer matrices use 8-bit affine quantization. Fusion projections, the final mixer, routers, norms, gates, and injection controls retain BF16.
 - The 51.2B-parameter PLE n-gram bank remains split into 128 shards. Shards 0 through 3 use 2-bit affine, group size 32. Shards 4 through 127 use 3-bit affine, group size 32.
-The importance matrix was collected from 1,024 tool-heavy multilingual calibration samples at sequence length 512. The included `oq_imatrix_report.json` records the allocation inputs and coverage.
 The MTP expert quantizer uses the calibrated activation importance of the analogous final backbone layer as a proxy. This preserves expert-specific weighting without claiming that the non-MTP calibration cache directly measured the draft head.
@@ -78,7 +78,7 @@ It installs the included model settings into a fresh oMLX base path, permits one
 ## Recommended generation settings
-Thinking mode is enabled by default. The upstream recommended settings are included in `generation_config.json`:
 ```text
 temperature: 1.0

 - MTP attention, shared-expert, and layer mixer matrices use 8-bit affine quantization. Fusion projections, the final mixer, routers, norms, gates, and injection controls retain BF16.
 - The 51.2B-parameter PLE n-gram bank remains split into 128 shards. Shards 0 through 3 use 2-bit affine, group size 32. Shards 4 through 127 use 3-bit affine, group size 32.
+The importance matrix was collected from 1,024 tool-heavy multilingual calibration samples at sequence length 512. It observed 73,665 of 73,728 backbone expert-projection slots. The 63 unobserved slots retained conservative fallback precision. The included `oq_imatrix_report.json` records the allocation inputs and complete coverage details.
 The MTP expert quantizer uses the calibrated activation importance of the analogous final backbone layer as a proxy. This preserves expert-specific weighting without claiming that the non-MTP calibration cache directly measured the draft head.
 ## Recommended generation settings
+Thinking mode is enabled by default. The tested generation settings are split between `generation_config.json` and the bundled `omlx_support/model_settings.json`:
 ```text
 temperature: 1.0

model.safetensors.index.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 94445280333
   },
   "weight_map": {
     "lm_head.biases": "model-00001-of-00018.safetensors",

 {
   "metadata": {
+    "total_size": 94444914776
   },
   "weight_map": {
     "lm_head.biases": "model-00001-of-00018.safetensors",