Instructions to use catid/MiniMax-M2.5-catid with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use catid/MiniMax-M2.5-catid with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="catid/MiniMax-M2.5-catid", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("catid/MiniMax-M2.5-catid", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("catid/MiniMax-M2.5-catid", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use catid/MiniMax-M2.5-catid with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "catid/MiniMax-M2.5-catid"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "catid/MiniMax-M2.5-catid",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/catid/MiniMax-M2.5-catid

SGLang

How to use catid/MiniMax-M2.5-catid with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "catid/MiniMax-M2.5-catid" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "catid/MiniMax-M2.5-catid",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "catid/MiniMax-M2.5-catid" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "catid/MiniMax-M2.5-catid",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use catid/MiniMax-M2.5-catid with Docker Model Runner:
```
docker model run hf.co/catid/MiniMax-M2.5-catid
```

MiniMax-M2.5-catid

Uncensored FP8 version of MiniMaxAI/MiniMax-M2.5 with safety refusal behavior removed via surgical weight replacement.

Refusal Removal Results

Evaluated on a 10,000-prompt refusal benchmark (8,000 train + 2,000 validation) using an LLM judge (GPT-5-nano) for 4-way classification (complied / refused / hedged / deflected):

Split	Total Prompts	Complied	Refused	Hedged	Deflected	Refusal Rate
Train	8,000	7,506	262	228	4	6.2%
Validation	2,000	1,885	55	59	1	5.8%

Coherence: 100% (50/50 capability test prompts answered correctly)

The ~6% residual "refusal rate" consists primarily of false positives from the LLM judge on benign prompts (opinion questions, casual banter, medical/privacy disclaimers) rather than actual safety refusals of harmful content.

Method

The o_proj (attention output projection) weights across all 62 transformer layers were replaced with weights from PRISM-PRO (an abliterated variant), dequantized from Q8_0 GGUF format and re-quantized to FP8 E4M3FN with block-wise scaling to match the original model's quantization scheme. All other weights (q_proj, k_proj, v_proj, MLP experts, embeddings, norms, etc.) are identical to the official FP8 base model.

Reconstruction error: 0.5% relative error per layer (cosine similarity ~1.0)
Modified weights: 62 o_proj tensors (3072 x 6144 each) + their scale_inv tensors
Unmodified weights: Everything else (~229B parameter MoE architecture preserved exactly)

Usage

This model is a drop-in replacement for MiniMaxAI/MiniMax-M2.5. Serve it with vLLM, SGLang, or any framework that supports the original model:

vLLM

vllm serve catid/MiniMax-M2.5-catid \
    --tensor-parallel-size 4 \
    --trust-remote-code \
    --max-model-len 2048

SGLang

python -m sglang.launch_server \
    --model catid/MiniMax-M2.5-catid \
    --tp 4 \
    --trust-remote-code

Recommended Parameters

temperature=1.0, top_p=0.95, top_k=40

Model Details

Architecture: MiniMax-M2.5 (229B MoE, 62 layers, 256 experts/layer, hidden_dim=3072)
Precision: FP8 E4M3FN with block-wise scaling (128x128 blocks)
Base model: MiniMaxAI/MiniMax-M2.5
Abliteration source: PrunaAI/MiniMax-M2.5-PRISM-PRO-Q8_0_v2-GGUF
License: Modified MIT (same as base model)

Disclaimer

This model is provided for research purposes. The removal of safety guardrails means it may generate content that the original model would refuse. Users are responsible for ensuring appropriate use.

Downloads last month: 15

Safetensors

Model size

229B params

Tensor type

F32

BF16

F8_E4M3

Model tree for catid/MiniMax-M2.5-catid

Base model

MiniMaxAI/MiniMax-M2.5

Quantized

(68)

this model