Instructions to use nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 with Transformers:

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

pipe = pipeline("text-generation", model="nameistoken/MiniMax-M2.5-Quark-W8A8-INT8", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nameistoken/MiniMax-M2.5-Quark-W8A8-INT8", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("nameistoken/MiniMax-M2.5-Quark-W8A8-INT8", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nameistoken/MiniMax-M2.5-Quark-W8A8-INT8

SGLang

How to use nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 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 "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 with Docker Model Runner:
```
docker model run hf.co/nameistoken/MiniMax-M2.5-Quark-W8A8-INT8
```

MiniMax-M2.5-Quark-W8A8-INT8

W8A8 INT8 quantized version of MiniMaxAI/MiniMax-M2.5 (456B MoE) using AMD Quark.

Model Details


Base Model	MiniMaxAI/MiniMax-M2.5
Architecture	MoE (Mixture of Experts), 62 layers, 256 experts, top-8 routing
Parameters	456B total, ~45.9B active
Quantization	W8A8 INT8 (per-channel weight + per-token dynamic activation)
Quantizer	AMD Quark (`ptpc_int8` scheme)
Model Size	216 GB (47 safetensors shards)
Original Size	~430 GB (BF16)
Compression	~2x size reduction

Quantization Scheme

Component	Dtype	Granularity	Mode
Weight	INT8	per-channel (ch_axis=0)	symmetric, static
Activation	INT8	per-token (ch_axis=1)	symmetric, dynamic
`lm_head`	BF16	—	unquantized
MoE gates	BF16	—	unquantized

How to Use

With vLLM (Recommended)

# Start the server
VLLM_WORKER_MULTIPROC_METHOD=spawn python -m vllm.entrypoints.openai.api_server \
    --model nameistoken/MiniMax-M2.5-Quark-W8A8-INT8 \
    --tensor-parallel-size 8 \
    --trust-remote-code \
    --max-model-len 8192

# Chat completion
curl http://localhost:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
  "model": "nameistoken/MiniMax-M2.5-Quark-W8A8-INT8",
  "messages": [{"role": "user", "content": "Hello! What is the capital of France?"}],
  "max_tokens": 256,
  "temperature": 0.7
}'

Hardware Requirements

Minimum: 8x GPUs with ≥48 GB VRAM each (e.g., AMD MI300X, AMD MI355X, NVIDIA A100-80G)
Tensor Parallelism: TP=8 required for 216 GB model

Quantization Details

This model was quantized using AMD Quark's ptpc_int8 (Per-Token Per-Channel INT8) scheme:

Weight quantization: INT8 per-channel (one scale per output channel), symmetric, static
Activation quantization: INT8 per-token (one scale per token), symmetric, dynamic (computed at inference time)
Excluded layers: lm_head (to preserve output quality) and all MoE gate layers (to preserve routing precision)

Reproduce Quantization

# Using AMD Quark
pip install quark transformers datasets torch

# Run quantization
cd Quark/examples/torch/language_modeling/llm_ptq
python quantize_quark_ptpc_int8.py \
    --model_dir /path/to/MiniMax-M2.5-bf16 \
    --quant_scheme ptpc_int8 \
    --num_calib_data 128 \
    --exclude_layers lm_head "*block_sparse_moe.gate*" \
    --skip_evaluation \
    --multi_gpu --multi_device \
    --trust_remote_code \
    --model_export hf_format \
    --output_dir /path/to/output

Citation

If you use this model, please cite the original MiniMax-M2.5 model:

@misc{minimax2025minimaxm25,
    title={MiniMax-M2.5},
    author={MiniMax},
    year={2025},
    url={https://huggingface.co/MiniMaxAI/MiniMax-M2.5}
}

License

This model inherits the Modified MIT License from the base model.

Downloads last month: 11

Safetensors

Model size

229B params

Tensor type

BF16

Model tree for nameistoken/MiniMax-M2.5-Quark-W8A8-INT8

Base model

MiniMaxAI/MiniMax-M2.5

Quantized

(68)

this model