Instructions to use QuantTrio/MiniMax-M2.1-AWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantTrio/MiniMax-M2.1-AWQ with Transformers:

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

pipe = pipeline("text-generation", model="QuantTrio/MiniMax-M2.1-AWQ", 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("QuantTrio/MiniMax-M2.1-AWQ", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("QuantTrio/MiniMax-M2.1-AWQ", 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 QuantTrio/MiniMax-M2.1-AWQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantTrio/MiniMax-M2.1-AWQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/MiniMax-M2.1-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantTrio/MiniMax-M2.1-AWQ

SGLang

How to use QuantTrio/MiniMax-M2.1-AWQ 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 "QuantTrio/MiniMax-M2.1-AWQ" \
    --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": "QuantTrio/MiniMax-M2.1-AWQ",
		"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 "QuantTrio/MiniMax-M2.1-AWQ" \
        --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": "QuantTrio/MiniMax-M2.1-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QuantTrio/MiniMax-M2.1-AWQ with Docker Model Runner:
```
docker model run hf.co/QuantTrio/MiniMax-M2.1-AWQ
```

Errors

by CihanDogan - opened Dec 29, 2025

Discussion

CihanDogan

Dec 29, 2025

Hi, with the latest version of vLLM (you specified), cuda 12.8 and 2x h100s I get the following error:

WorkerProc failed to start.
Traceback (most recent call last):
File "/root/.venv/lib/python3.12/site-packages/vllm/v1/executor/multiproc_executor.py", line 722, in worker_main
worker = WorkerProc(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/.venv/lib/python3.12/site-packages/vllm/v1/executor/multiproc_executor.py", line 562, in init
self.worker.load_model()
File "/root/.venv/lib/python3.12/site-packages/vllm/v1/worker/gpu_worker.py", line 289, in load_model
self.model_runner.load_model(eep_scale_up=eep_scale_up)
File "/root/.venv/lib/python3.12/site-packages/vllm/v1/worker/gpu_model_runner.py", line 3581, in load_model
self.model = model_loader.load_model(
^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/.venv/lib/python3.12/site-packages/vllm/model_executor/model_loader/base_loader.py", line 56, in load_model
process_weights_after_loading(model, model_config, target_device)
File "/root/.venv/lib/python3.12/site-packages/vllm/model_executor/model_loader/utils.py", line 108, in process_weights_after_loading
quant_method.process_weights_after_loading(module)
File "/root/.venv/lib/python3.12/site-packages/vllm/model_executor/layers/quantization/awq_marlin.py", line 585, in process_weights_after_loading
marlin_w13_qweight = ops.awq_marlin_moe_repack(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/.venv/lib/python3.12/site-packages/vllm/_custom_ops.py", line 1327, in awq_marlin_moe_repack
output[e] = torch.ops._C.awq_marlin_repack(
~~~~~~^^^

torch.AcceleratorError: CUDA error: the provided PTX was compiled with an unsupported toolchain.
Search for cudaErrorUnsupportedPtxVersion' in https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html for more information. CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with TORCH_USE_CUDA_DSA` to enable device-side assertions.

CihanDogan

Dec 29, 2025

From nvidia-smi:
Driver: 570.195.03 ✅
CUDA: 12.8 ✅
GPUs: 2× H100 PCIe (sm_90a) ✅

CihanDogan

Dec 29, 2025

Hi, the error was on my end, this fixed it: https://github.com/vllm-project/vllm/issues/31027

mtcl

Dec 29, 2025

I was about to say. It works perfectly in my side. Better than cyankiwi quants!

CihanDogan changed discussion status to closed Dec 29, 2025

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment