---
license: other
base_model: MiniMaxAI/MiniMax-M3
base_model_relation: finetune
library_name: transformers
pipeline_tag: text-generation
language: [en]
tags:
  - uncensored
  - abliterated
  - moe
  - minimax
  - security
---

![MiniMax-M3-uncensored](banner.png)

# MiniMax-M3-uncensored (BF16)

**TL;DR:** an uncensored build of [MiniMaxAI/MiniMax-M3](https://huggingface.co/MiniMaxAI/MiniMax-M3)
(428B-parameter MoE, 23B active), with **0 hard refusals** on harmful prompts while the model's
capabilities stay intact. Full-precision BF16, 796 GB.

> ⚠️ **This model is genuinely uncensored, it will comply with requests a stock model refuses.**
>
> **Intended use, the constructive side.** An assistant that does not refuse is genuinely useful for
> **ethical hacking, security research, and penetration testing**: red-teaming, analyzing malware and
> exploit code, writing detection/YARA rules, reviewing vulnerabilities, and studying attack
> techniques without the model bailing out mid-task. Use it lawfully and responsibly, you are
> accountable for what you do with it.

## Facts & figures

| | |
|---|---|
| Base model | [MiniMaxAI/MiniMax-M3](https://huggingface.co/MiniMaxAI/MiniMax-M3) (428B total / 23B active, MoE, multimodal, 1M context) |
| Precision | BF16 (796 GB, 59 shards) |
| Effective refusals | **0/16 hard-refusals** on harmful prompts (mlabonne/harmful_behaviors); the base model deliberates or refuses |
| Modified weights | attention `o_proj` + every residual-writing `down_proj` (dense, shared expert, and all 128 routed experts per MoE layer), all 60 layers |
| Coherence | intact (multimodal, reasoning and MoE routing preserved) |

## Uncensoring, verified

Generation on harmful prompts, hard-refusal phrases only ("I cannot" / "I won't" etc.):

| Prompt set | Prompts | Hard refusals |
|---|---:|:---:|
| mlabonne/harmful_behaviors | 16 | **0/16 (0.0%)** |

The abliteration is a **weight property**: it survives quantization, so downstream NVFP4 / GGUF builds
keep the same behavior.

> ℹ️ **MiniMax-M3 has a reasoning mode** (`<mm:think>`). The model thinks before answering; the
> uncensored build reasons about *how* to fulfill a request rather than whether to refuse. For direct
> answers, disable thinking in your client.

## Run it with transformers

```python
from transformers import AutoModelForImageTextToText, AutoTokenizer, AutoConfig
tok = AutoTokenizer.from_pretrained("ressl/MiniMax-M3-uncensored")
cfg = AutoConfig.from_pretrained("ressl/MiniMax-M3-uncensored")
model = AutoModelForImageTextToText.from_pretrained(
    "ressl/MiniMax-M3-uncensored", config=cfg, dtype="bfloat16", device_map="auto")
```

## Run it with vLLM

```bash
vllm serve ressl/MiniMax-M3-uncensored \
  --tensor-parallel-size 8 --tool-call-parser minimax_m3 --reasoning-parser minimax_m3 --trust-remote-code
```

MiniMax-M3 needs a recent vLLM with M3 support (for RTX PRO 6000 / Blackwell see
[0xSero/minimax-m3-sm120](https://github.com/0xSero/minimax-m3-sm120)). The 428B MoE needs multi-GPU
at BF16; an **NVFP4 quant** (NVIDIA ModelOpt, following
[nvidia/MiniMax-M3-NVFP4](https://huggingface.co/nvidia/MiniMax-M3-NVFP4)) shrinks it to ~230 GB and
runs on a single Blackwell node.

## Quality & limitations

- 0/16 hard refusals on a harmful-prompt sample; not a full capability benchmark.
- BF16 is large (796 GB); for single-node serving, produce an NVFP4 quant via NVIDIA's ModelOpt
  recipe on this checkpoint (the abliteration is a weight property and survives quantization).
- The reasoning mode is on by default (see the note above).

## ❤️ Support

Producing and validating an uncensored build of a brand-new 428B MoE was a **lot of work**. If it's
useful to you, I'd genuinely appreciate your support on [**Patreon**](https://www.patreon.com/cw/ressl) 🙏,
more at [ressl.ch](https://ressl.ch/).

## License & credits

License **inherited from the base model** by **MiniMaxAI**. Uncensoring and validation by
**Robert Ressl** ([Hugging Face](https://huggingface.co/ressl) · [Website](https://ressl.ch/) · [LinkedIn](https://www.linkedin.com/in/robertressl/) · [Patreon](https://www.patreon.com/cw/ressl)).