Instructions to use MiniMaxAI/MiniMax-M1-40k-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MiniMaxAI/MiniMax-M1-40k-hf with Transformers:

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

pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-M1-40k-hf")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MiniMaxAI/MiniMax-M1-40k-hf")
model = AutoModelForCausalLM.from_pretrained("MiniMaxAI/MiniMax-M1-40k-hf")
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 MiniMaxAI/MiniMax-M1-40k-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MiniMaxAI/MiniMax-M1-40k-hf"
# 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-M1-40k-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MiniMaxAI/MiniMax-M1-40k-hf

SGLang

How to use MiniMaxAI/MiniMax-M1-40k-hf 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-M1-40k-hf" \
    --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-M1-40k-hf",
		"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-M1-40k-hf" \
        --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-M1-40k-hf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MiniMaxAI/MiniMax-M1-40k-hf with Docker Model Runner:
```
docker model run hf.co/MiniMaxAI/MiniMax-M1-40k-hf
```

QscQ commited on Jul 3, 2025

Commit

3d2b70f

verified ·

1 Parent(s): 61d2a4f

Update config.json

Browse files

Files changed (1) hide show

config.json +90 -95

config.json CHANGED Viewed

@@ -1,94 +1,90 @@
 {
   "architectures": [
-    "MiniMaxM1ForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "attn_type_list": [
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    0,
-    1
   ],
-  "auto_map": {
-    "AutoConfig": "configuration_minimax_m1.MiniMaxM1Config",
-    "AutoModelForCausalLM": "modeling_minimax_m1.MiniMaxM1ForCausalLM"
-  },
   "bos_token_id": null,
   "eos_token_id": null,
   "head_dim": 128,
@@ -96,14 +92,14 @@
   "hidden_size": 6144,
   "initializer_range": 0.02,
   "intermediate_size": 9216,
-  "layernorm_full_attention_alpha": 3.5565588200778455,
-  "layernorm_full_attention_beta": 1.0,
-  "layernorm_linear_attention_alpha": 3.5565588200778455,
-  "layernorm_linear_attention_beta": 1.0,
-  "layernorm_mlp_alpha": 3.5565588200778455,
-  "layernorm_mlp_beta": 1.0,
   "max_position_embeddings": 10240000,
-  "model_type": "minimax_m1",
   "num_attention_heads": 64,
   "num_experts_per_tok": 2,
   "num_hidden_layers": 80,
@@ -123,5 +119,4 @@
   "transformers_version": "4.45.2",
   "use_cache": true,
   "vocab_size": 200064
-}

 {
   "architectures": [
+    "MiniMaxForCausalLM"
   ],
   "attention_dropout": 0.0,
+  "layer_types": [
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "linear_attention",
+    "full_attention"
   ],
   "bos_token_id": null,
   "eos_token_id": null,
   "head_dim": 128,
   "hidden_size": 6144,
   "initializer_range": 0.02,
   "intermediate_size": 9216,
+  "full_attn_alpha_factor": 3.5565588200778455,
+  "full_attn_beta_factor": 1.0,
+  "linear_attn_alpha_factor": 3.5565588200778455,
+  "linear_attn_beta_factor": 1.0,
+  "mlp_alpha_factor": 3.5565588200778455,
+  "mlp_beta_factor": 1.0,
   "max_position_embeddings": 10240000,
+  "model_type": "minimax",
   "num_attention_heads": 64,
   "num_experts_per_tok": 2,
   "num_hidden_layers": 80,
   "transformers_version": "4.45.2",
   "use_cache": true,
   "vocab_size": 200064
+}