MiniMax-M2.7-abliterated-Heretic-FP8

Instructions to use dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8", 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("dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8", 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 dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8

SGLang

How to use dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8 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 "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8" \
    --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": "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8",
		"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 "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8" \
        --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": "dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8 with Docker Model Runner:
```
docker model run hf.co/dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8
```

MiniMax-M2.7 Abliterated Heretic — FP8

FP8 dynamic (per-channel weight, per-token activation) quantization of Youssofal/MiniMax-M2.7-abliterated-BF16, which is itself a Heretic-method abliteration of MiniMaxAI/MiniMax-M2.7.

Lineage

Base model: MiniMaxAI/MiniMax-M2.7 (229B MoE, 10B active, 256 experts, 200K ctx)
Abliteration: Youssofal/MiniMax-M2.7-abliterated-BF16 — Heretic / Ablated Refusal Adaptation (ARA)
This repo: FP8 quant of the above for fast vLLM inference

Format

Weights: float8_e4m3fn, per-output-channel symmetric scales (float32)
Activations: dynamic per-token FP8 at runtime
KV cache: run with --kv-cache-dtype fp8 for full FP8 serving
Config: compressed-tensors / format: float-quantized, ignored: lm_head
Tensors: 96,165 total, 47 safetensors shards, ~230 GB

Serve with vLLM

vllm serve LittleNicky55/MiniMax-M2.7-abliterated-Heretic-FP8 \
  --tensor-parallel-size 2 \
  --dtype bfloat16 \
  --kv-cache-dtype fp8 \
  --max-model-len 196608 \
  --gpu-memory-utilization 0.92 \
  --trust-remote-code \
  --enable-prefix-caching

Fits comfortably in 2× H200 141GB (total VRAM budget ~230 GB + KV + compute).

Quantization method

Streaming per-shard quantization script: for each Linear weight W, compute per-output-channel scale = |W|.amax(dim=1) / 448.0, then W_fp8 = (W / scale).to(fp8_e4m3fn). No calibration data required (FP8_DYNAMIC scheme).

License

Inherits the non-commercial MiniMax M-Series license from the base model.

Downloads last month: 16

Safetensors

Model size

229B params

Tensor type

F32

F16

F8_E4M3

BF16

Model tree for dokjlahdf/MiniMax-M2.7-abliterated-Heretic-FP8

Base model

MiniMaxAI/MiniMax-M2.7

Finetuned

Youssofal/MiniMax-M2.7-abliterated-BF16

Quantized

(9)

this model