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

Libraries

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

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

pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-M2.1", 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("MiniMaxAI/MiniMax-M2.1", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MiniMaxAI/MiniMax-M2.1", 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]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MiniMaxAI/MiniMax-M2.1 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/MiniMaxAI/MiniMax-M2.1

SGLang

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

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

xuebi commited on Dec 26, 2025

Commit

2713e50

1 Parent(s): 1ccbe32

update

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -109,7 +109,7 @@ Furthermore, across specific benchmarks—including test case generation, code p
 | SWE-Review | 8.9 | 3.4 | 10.5 | 16.2 | x | x | 6.4 |
 | OctoCodingbench | 26.1 | 13.3 | 22.8 | 36.2 | 22.9 | x | 26.0 |
-To evaluate the model's full-stack capability to architect complete, functional applications "from zero to one," we established a novel benchmark: [VIBE (Visual & Interactive Benchmark for Execution)](https://huggingface.co/datasets/MiniMaxAI/VIBE). This suite encompasses five core subsets: Web, Simulation, Android, iOS, and Backend. Distinguishing itself from traditional benchmarks, VIBE leverages an innovative Agent-as-a-Verifier (AaaV) paradigm to automatically assess the interactive logic and visual aesthetics of generated applications within a real runtime environment.
 MiniMax-M2.1 delivers outstanding performance on the VIBE aggregate benchmark, achieving an average score of 88.6—demonstrating robust full-stack development capabilities. It excels particularly in the VIBE-Web (91.5) and VIBE-Android (89.7) subsets.
@@ -171,7 +171,7 @@ We recommend using [Transformers](https://github.com/huggingface/transformers) t
 ### Other Inference Engines
-- [KTransformers](https://github.com/kvcache-ai/ktransformers)
 ### Inference Parameters

 | SWE-Review | 8.9 | 3.4 | 10.5 | 16.2 | x | x | 6.4 |
 | OctoCodingbench | 26.1 | 13.3 | 22.8 | 36.2 | 22.9 | x | 26.0 |
+To evaluate the model's full-stack capability to architect complete, functional applications "from zero to one," we established a novel benchmark: [VIBE (Visual & Interactive Benchmark for Execution in Application Development)](https://huggingface.co/datasets/MiniMaxAI/VIBE). This suite encompasses five core subsets: Web, Simulation, Android, iOS, and Backend. Distinguishing itself from traditional benchmarks, VIBE leverages an innovative Agent-as-a-Verifier (AaaV) paradigm to automatically assess the interactive logic and visual aesthetics of generated applications within a real runtime environment.
 MiniMax-M2.1 delivers outstanding performance on the VIBE aggregate benchmark, achieving an average score of 88.6—demonstrating robust full-stack development capabilities. It excels particularly in the VIBE-Web (91.5) and VIBE-Android (89.7) subsets.
 ### Other Inference Engines
+- [KTransformers](https://github.com/kvcache-ai/ktransformers/blob/main/doc/en/kt-kernel/MiniMax-M2.1-Tutorial.md)
 ### Inference Parameters