Instructions to use Youssofal/Qwen3.6-27B-MTPLX-Optimized with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Youssofal/Qwen3.6-27B-MTPLX-Optimized 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("Youssofal/Qwen3.6-27B-MTPLX-Optimized")

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
LM Studio

Pi new

How to use Youssofal/Qwen3.6-27B-MTPLX-Optimized with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "Youssofal/Qwen3.6-27B-MTPLX-Optimized"

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": "Youssofal/Qwen3.6-27B-MTPLX-Optimized"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use Youssofal/Qwen3.6-27B-MTPLX-Optimized 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 "Youssofal/Qwen3.6-27B-MTPLX-Optimized"

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 Youssofal/Qwen3.6-27B-MTPLX-Optimized

Run Hermes

hermes

MLX LM

How to use Youssofal/Qwen3.6-27B-MTPLX-Optimized with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "Youssofal/Qwen3.6-27B-MTPLX-Optimized"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "Youssofal/Qwen3.6-27B-MTPLX-Optimized"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "Youssofal/Qwen3.6-27B-MTPLX-Optimized",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Youssofal commited on 19 days ago

Commit

aed68a8

verified ·

1 Parent(s): ac8e4d3

Drop remaining 'coming soon' references; MTPLX is released

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -40,9 +40,9 @@ mtplx start
 This artifact pairs the Qwen3.6-27B trunk — MLX-quantized with MTPLX's `gdn8-speed4` policy (8-bit Gated Delta Network linears, 4-bit MLP, BF16 norms) — with a **calibrated INT4 Multi-Token-Prediction sidecar** grafted onto the trunk. The MTP head is what enables *native* speculative decoding: the model drafts its own tokens, with no external draft model required.
-When MTPLX is released, it will accept those draft tokens with **mathematically exact** probability-ratio acceptance and residual correction, so the speculative path stays distribution-preserving at realistic coding settings (`temperature=0.6`, `top_p=0.95`, `top_k=20`) — not just greedy.
-Until then you can still:
 - Inspect the architecture and MTP tensors with any `safetensors` reader.
 - Use the trunk weights with [`mlx-lm`](https://github.com/ml-explore/mlx-lm) for ordinary autoregressive decoding (the MTP head is sidecar-only and ignored by `mlx-lm`).
@@ -104,5 +104,5 @@ This checkpoint is released under the **Apache License 2.0**, matching the Qwen3
 ## Links
-- **Runtime (coming soon)**: [github.com/youssofal/mtplx](https://github.com/youssofal/mtplx)
 - **Base model**: [Qwen/Qwen3.6-27B](https://huggingface.co/Qwen/Qwen3.6-27B)

 This artifact pairs the Qwen3.6-27B trunk — MLX-quantized with MTPLX's `gdn8-speed4` policy (8-bit Gated Delta Network linears, 4-bit MLP, BF16 norms) — with a **calibrated INT4 Multi-Token-Prediction sidecar** grafted onto the trunk. The MTP head is what enables *native* speculative decoding: the model drafts its own tokens, with no external draft model required.
+MTPLX accepts those draft tokens with **mathematically exact** probability-ratio acceptance and residual correction, so the speculative path stays distribution-preserving at realistic coding settings (`temperature=0.6`, `top_p=0.95`, `top_k=20`) — not just greedy.
+You can also:
 - Inspect the architecture and MTP tensors with any `safetensors` reader.
 - Use the trunk weights with [`mlx-lm`](https://github.com/ml-explore/mlx-lm) for ordinary autoregressive decoding (the MTP head is sidecar-only and ignored by `mlx-lm`).
 ## Links
+- **Runtime**: [github.com/youssofal/MTPLX](https://github.com/youssofal/MTPLX)  ·  `pip install mtplx`
 - **Base model**: [Qwen/Qwen3.6-27B](https://huggingface.co/Qwen/Qwen3.6-27B)