Instructions to use FlatFootInternational/qwen3.8-27b-MTPLX-5bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FlatFootInternational/qwen3.8-27b-MTPLX-5bit 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("FlatFootInternational/qwen3.8-27b-MTPLX-5bit")

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 FlatFootInternational/qwen3.8-27b-MTPLX-5bit with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"

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": "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use FlatFootInternational/qwen3.8-27b-MTPLX-5bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "FlatFootInternational/qwen3.8-27b-MTPLX-5bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use FlatFootInternational/qwen3.8-27b-MTPLX-5bit 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 "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"

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 FlatFootInternational/qwen3.8-27b-MTPLX-5bit

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use FlatFootInternational/qwen3.8-27b-MTPLX-5bit with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "FlatFootInternational/qwen3.8-27b-MTPLX-5bit"

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 "FlatFootInternational/qwen3.8-27b-MTPLX-5bit" \
  --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"

Qwen 3.8 27B MTPLX 5-bit

Every weight matrix at 5-bit, sensitive parts at 16-bit, native multi-token-prediction head kept, so MTPLX still drafts ahead and verifies in one pass.

Speeds

Measured on an Macbook M5 (NOT pro or max) with 32GB unified memory, fans verified at max, single stream, generation running to the model's own stop, official Qwen 3.8 sampling (temperature 1.0, top-p 0.95, top-k 20).

How it is built

Every weight matrix at 5-bit with 64-weight groups.
The GDN convolution kernels and recurrent state parameters, every norm, and the whole MTP head stay 16-bit.
Download 19.4 GB
Context window 262,144 tokens
MTP depth 3
Sampling: temperature 1.0, top-p 0.95, top-k 20 (the official Qwen 3.8 contract)

Use it

You want 32 GB of unified memory or more for this one.

Command line:

pip install mtplx

mtplx serve --model FlatFootInternational/qwen3.8-27b-MTPLX-5bit

Downloads last month: 292

Safetensors

Model size

5B params

Tensor type

BF16

U32

MLX

Hardware compatibility

4-bit

Model tree for FlatFootInternational/qwen3.8-27b-MTPLX-5bit

Base model

Qwen/Qwen3.8-27B

Quantized

(935)

this model