Instructions to use a-m-team/AM-Thinking-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use a-m-team/AM-Thinking-v1 with Transformers:

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

pipe = pipeline("text-generation", model="a-m-team/AM-Thinking-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("a-m-team/AM-Thinking-v1")
model = AutoModelForCausalLM.from_pretrained("a-m-team/AM-Thinking-v1")
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
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use a-m-team/AM-Thinking-v1 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/a-m-team/AM-Thinking-v1

SGLang

How to use a-m-team/AM-Thinking-v1 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 "a-m-team/AM-Thinking-v1" \
    --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": "a-m-team/AM-Thinking-v1",
		"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 "a-m-team/AM-Thinking-v1" \
        --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": "a-m-team/AM-Thinking-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use a-m-team/AM-Thinking-v1 with Docker Model Runner:
```
docker model run hf.co/a-m-team/AM-Thinking-v1
```

AIME 2024 contamination

by autoprogrammer - opened May 24, 2025

Discussion

autoprogrammer

May 24, 2025

Why the AIME2024 is in the training data and you also report it as benchmark in your paper?

Emperorizzis

am team org May 25, 2025

Hello, we implemented strict data decontamination strategies during both the SFT and GRPO stages, including:
1. Exact decontamination
2. Semantic decontamination

We guarantee that none of our training data contains any AIME2024 data.

Emperorizzis changed discussion status to closed May 25, 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