Instructions to use sailing-lab/SR2AM-v1.0-30B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sailing-lab/SR2AM-v1.0-30B with Transformers:

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

pipe = pipeline("text-generation", model="sailing-lab/SR2AM-v1.0-30B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sailing-lab/SR2AM-v1.0-30B")
model = AutoModelForCausalLM.from_pretrained("sailing-lab/SR2AM-v1.0-30B")
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 sailing-lab/SR2AM-v1.0-30B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sailing-lab/SR2AM-v1.0-30B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sailing-lab/SR2AM-v1.0-30B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sailing-lab/SR2AM-v1.0-30B

SGLang

How to use sailing-lab/SR2AM-v1.0-30B 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 "sailing-lab/SR2AM-v1.0-30B" \
    --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": "sailing-lab/SR2AM-v1.0-30B",
		"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 "sailing-lab/SR2AM-v1.0-30B" \
        --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": "sailing-lab/SR2AM-v1.0-30B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sailing-lab/SR2AM-v1.0-30B with Docker Model Runner:
```
docker model run hf.co/sailing-lab/SR2AM-v1.0-30B
```

mingkaid commited on 4 days ago

Commit

dce345c

verified ·

1 Parent(s): ee4a2f3

Fill GitHub URL in model card

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -22,7 +22,7 @@ We argue that efficient agentic reasoning benefits from decomposing deliberation
 SR²AM-v1.0-30B achieves an overall Pass@1 of **71.3** across 11 benchmarks spanning math, science, tabular analysis, and web information seeking — competitive with systems at 685B–1T parameters, while consuming **25–95% fewer reasoning tokens** than comparably sized agentic LLMs.
-More details: [project website](https://sr2am-agentic-llm.github.io/) | [paper](https://arxiv.org) | [GitHub](<GITHUB_URL_TBD>).
 ## Key Features
@@ -34,7 +34,7 @@ More details: [project website](https://sr2am-agentic-llm.github.io/) | [paper](
 ## Quick Start
-See the [GitHub repository](<GITHUB_URL_TBD>) for setup and inference instructions.
 ## Main Results

 SR²AM-v1.0-30B achieves an overall Pass@1 of **71.3** across 11 benchmarks spanning math, science, tabular analysis, and web information seeking — competitive with systems at 685B–1T parameters, while consuming **25–95% fewer reasoning tokens** than comparably sized agentic LLMs.
+More details: [project website](https://sr2am-agentic-llm.github.io/) | [paper](https://arxiv.org) | [GitHub](https://github.com/sailing-lab/sr2am).
 ## Key Features
 ## Quick Start
+See the [GitHub repository](https://github.com/sailing-lab/sr2am) for setup and inference instructions.
 ## Main Results