---
title: MiMo-V2.6-Pro-RL-mxfp4-q8
canonical_url: "https://www.modelscope.cn/models/mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8"
md_url: "https://www.modelscope.cn/models/mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8.md"
repository: mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8
last_updated: 2026-09-22
license: mit
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - mimo_v2
architectures:
  - MiMoV2ForCausalLM
base_model:
  - XiaomiMiMo/MiMo-V2.6-Pro-RL
base_model_relation: quantized
parameters: 133.5B
tensor_type:
  - BF16
  - U32
  - F32
  - U8
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 7
stars: 0
tags:
  - mlx
  - mimo_v2
  - agent
  - long-context
---

# MiMo-V2.6-Pro-RL-mxfp4-q8

> MiMo-V2.6-Pro-RL-mxfp4-q8 - mlx-community 在 ModelScope 开源的模型。mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8

mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8 是 ModelScope 魔搭社区上的 133.5B 参数text-generation模型，采用 mit 许可，基于 XiaomiMiMo/MiMo-V2.6-Pro-RL 构建。

- **Repository**: mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8
- **License**: mit
- **Tasks**: text-generation
- **Parameters**: 133.5B
- **Base model**: XiaomiMiMo/MiMo-V2.6-Pro-RL
- **Tags**: mlx, mimo_v2, agent, long-context
- **Downloads**: 7
- **Stars**: 0
- **Last updated**: 2026-09-22

Source: https://www.modelscope.cn/models/mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8

---

# mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8

This model [mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8](https://huggingface.co/mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8) was
converted to MLX format from [XiaomiMiMo/MiMo-V2.6-Pro-RL](https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-RL)
using mlx-lm version **0.32.0** ([PR #1219](https://github.com/ml-explore/mlx-lm/pull/1219)).

## Quantization

- MoE expert weights are the original checkpoint's native **MXFP4** (4-bit, group size 32), loaded directly without requantization.
- Attention, dense MLP, embeddings and `lm_head` are **8-bit affine**, group size 64.
- 4.339 bits per weight overall, 516 GB on disk.

This is a text-only conversion: the vision and audio encoders and the MTP/DFlash draft weights are not included.

## Requirements

MiMo-V2 support is in [mlx-lm PR #1219](https://github.com/ml-explore/mlx-lm/pull/1219). Until it is merged, install mlx-lm from that branch:

```bash
pip install git+https://github.com/kernelpool/mlx-lm.git@add-mimo-v2
```

At 1.02T total parameters (42B active) this model does not fit on a single Mac. It is meant to run tensor-parallel across two 512 GB machines with mlx-lm's distributed support, for example:

```bash
mlx.launch --backend jaccl --hostfile hosts.json -- \
  python -m mlx_lm.examples.sharded_generate \
  --model mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8 --prompt "hello" -m 256
```

See the mlx-lm distributed inference documentation for the hostfile format and backend setup.

## Use with mlx

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/MiMo-V2.6-Pro-RL-mxfp4-q8")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True, return_dict=False,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```

Thinking is enabled by default in the chat template; pass `enable_thinking=False` to `apply_chat_template` to disable it. Tool calls use the Qwen3-Coder format (`<tool_call><function=...>`), which the `qwen3_coder` tool parser in mlx-lm handles.