Instructions to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with MLX:

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

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

model, tokenizer = load("xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp")

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

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

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"

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": "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

MLX LM

How to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Hermes Agent

How to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp with OpenClaw:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp"

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp" \
  --custom-provider-id mlx-lm \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

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

MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp

An Apple MLX enhanced oQ5 (oQ5e) quantization of XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B, built with oMLX using 128 calibration samples at sequence length 512, strict importance-matrix coverage, BF16 compute, and group size 64.

Key Configuration

Modality: Pure Text-Only (LLM). Vision tower weights and multimodal processor configs have been cleanly omitted, saving ~1.5 GB of unified memory.
Speculative Decoding (MTP): Includes the official 15-tensor Multi-Token Prediction draft head (model-mtp.safetensors, grafted from Qwen/Qwen3.5-9B). When loaded in oMLX with mtp_enabled: true, speculative decoding achieves ~1.4x–1.6x faster token generation on text tasks.

Validation

Validated locally before upload:

Strict model and MTP module instantiation through the oMLX runtime
Global 5-bit affine quantization metadata with group size 64
Complete strict oQe importance-matrix application
Tokenizer SHA-256 identity verified against official source
Official chat template renders OpenAI-style function schemas and thinking blocks cleanly
Deterministic generation smoke test verified

Usage with oMLX

To enable speculative decoding acceleration:

// ~/.omlx/model_settings.json
"MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp": {
  "mtp_enabled": true,
  "mtp_num_draft_tokens": 3,
  "temperature": 0.6,
  "repetition_penalty": 1.05
}

License

MIT, following the upstream model.

Downloads last month: 277

Safetensors

Model size

9B params

Tensor type

U32

BF16

MLX

Hardware compatibility

5-bit

Model tree for xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Text-oQ5e-mtp

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Finetuned

XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B

Quantized

(56)

this model