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

Libraries

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

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

pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-M2.5", 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-M2.5", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MiniMaxAI/MiniMax-M2.5", 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-M2.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MiniMaxAI/MiniMax-M2.5"
# 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-M2.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

SGLang

How to use MiniMaxAI/MiniMax-M2.5 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-M2.5" \
    --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-M2.5",
		"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-M2.5" \
        --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-M2.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

Update README.md

#29

by areelkhairil - opened Feb 19

base: refs/heads/main

←

from: refs/pr/29

Discussion Files changed

-1

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -264,7 +264,7 @@ Evaluation methods:
 > - Terminal Bench 2: We tested Terminal Bench 2 using Claude Code 2.0.64 as the evaluation scaffolding. We modified the Dockerfiles of some problems to ensure the correctness of the problems themselves, uniformly expanded sandbox specifications to 8-core CPU and 16 GB memory, set the timeout uniformly to 7,200 seconds, and equipped each problem with a basic toolset (ps, curl, git, etc.). While not retrying on timeouts, we added a detection mechanism for empty scaffolding responses, retrying tasks whose final response was empty to handle various abnormal interruption scenarios. Final results are averaged over 4 runs.
 > - VIBE-Pro: Internal benchmark. Uses Claude Code as the scaffolding to automatically verify the interaction logic and visual effects of programs. All scores are computed through a unified pipeline that includes a requirements set, containerized deployment, and a dynamic interaction environment. Final results are averaged over 3 runs.
 > - BrowseComp: Uses the same agent framework as WebExplorer (Liu et al., 2025). When token usage exceeds 30% of the maximum context, all history is discarded.
-> - Wide Search: Uses the same agent framework as WebExplorer (Liu et al., 2025).
 > - RISE: Internal benchmark. Contains real questions from human experts, evaluating the model's multi-step information retrieval and reasoning capabilities when combined with complex web interactions. A Playwright-based browser tool suite is added on top of the WebExplorer (Liu et al., 2025) agent framework.
 > - GDPval-MM: Internal benchmark. Based on the open-source GDPval test set, using a custom agentic evaluation framework where an LLM-as-a-judge performs pairwise win/tie/loss judgments on complete trajectories. Average token cost per task is calculated based on each vendor's official API pricing (without caching).
 > - MEWC: Internal benchmark. Built on MEWC (Microsoft Excel World Championship), comprising 179 problems from the main and other regional divisions of Excel esports competitions from 2021–2026. It evaluates the model's ability to understand competition Excel spreadsheets and use Excel tools to complete problems. Scores are calculated by comparing output and answer cell values one by one.

 > - Terminal Bench 2: We tested Terminal Bench 2 using Claude Code 2.0.64 as the evaluation scaffolding. We modified the Dockerfiles of some problems to ensure the correctness of the problems themselves, uniformly expanded sandbox specifications to 8-core CPU and 16 GB memory, set the timeout uniformly to 7,200 seconds, and equipped each problem with a basic toolset (ps, curl, git, etc.). While not retrying on timeouts, we added a detection mechanism for empty scaffolding responses, retrying tasks whose final response was empty to handle various abnormal interruption scenarios. Final results are averaged over 4 runs.
 > - VIBE-Pro: Internal benchmark. Uses Claude Code as the scaffolding to automatically verify the interaction logic and visual effects of programs. All scores are computed through a unified pipeline that includes a requirements set, containerized deployment, and a dynamic interaction environment. Final results are averaged over 3 runs.
 > - BrowseComp: Uses the same agent framework as WebExplorer (Liu et al., 2025). When token usage exceeds 30% of the maximum context, all history is discarded.
+> - Wide Search: Uses the same agent framework as WebExplorer (Liu et al., 2025)..
 > - RISE: Internal benchmark. Contains real questions from human experts, evaluating the model's multi-step information retrieval and reasoning capabilities when combined with complex web interactions. A Playwright-based browser tool suite is added on top of the WebExplorer (Liu et al., 2025) agent framework.
 > - GDPval-MM: Internal benchmark. Based on the open-source GDPval test set, using a custom agentic evaluation framework where an LLM-as-a-judge performs pairwise win/tie/loss judgments on complete trajectories. Average token cost per task is calculated based on each vendor's official API pricing (without caching).
 > - MEWC: Internal benchmark. Built on MEWC (Microsoft Excel World Championship), comprising 179 problems from the main and other regional divisions of Excel esports competitions from 2021–2026. It evaluates the model's ability to understand competition Excel spreadsheets and use Excel tools to complete problems. Scores are calculated by comparing output and answer cell values one by one.