Instructions to use WeiboAI/VibeThinker-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WeiboAI/VibeThinker-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="WeiboAI/VibeThinker-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("WeiboAI/VibeThinker-1.5B")
model = AutoModelForCausalLM.from_pretrained("WeiboAI/VibeThinker-1.5B", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use WeiboAI/VibeThinker-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WeiboAI/VibeThinker-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WeiboAI/VibeThinker-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/WeiboAI/VibeThinker-1.5B

SGLang

How to use WeiboAI/VibeThinker-1.5B 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 "WeiboAI/VibeThinker-1.5B" \
    --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": "WeiboAI/VibeThinker-1.5B",
		"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 "WeiboAI/VibeThinker-1.5B" \
        --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": "WeiboAI/VibeThinker-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use WeiboAI/VibeThinker-1.5B with Docker Model Runner:
```
docker model run hf.co/WeiboAI/VibeThinker-1.5B
```

YinZhiBin commited on Nov 11, 2025

Commit

863a719

verified ·

1 Parent(s): f97107b

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -2

README.md CHANGED Viewed

@@ -26,7 +26,7 @@ VibeThinker-1.5B is a 1.5-billion parameter dense language model. With a total t
 🌱 Code Generation: It achieved scores of 55.9 on LiveCodeBench v5 and 51.1 on v6. Its v6 score slightly leads Magistral Medium (50.3), underscoring its strong reasoning performance.
-![image](https://cdn-uploads.huggingface.co/production/uploads/64d1faaa1ed6649d70d1fa2f/JZ7eu98mpxMQHHlTB_Uah.png)
 🔁 On the AIME 25 benchmark, VibeThinker-1.5B significantly extends the Pareto frontier of reasoning accuracy versus model scale, demonstrating that exceptional performance can be achieved with extreme parameter efficiency.
@@ -34,7 +34,9 @@ VibeThinker-1.5B is a 1.5-billion parameter dense language model. With a total t
 ## Training Pipeline
-![image](https://cdn-uploads.huggingface.co/production/uploads/64d1faaa1ed6649d70d1fa2f/9d2--eHAW3aOlvdTe48cR.png)
 VibeThinker-1.5B's core innovation lies in the "Spectrum-to-Signal Principle" (SSP) training framework: it first explores solution diversity during the Supervised Fine-Tuning (SFT) stage, and then optimizes its policy to reinforce correct signals in the Reinforcement Learning (RL) stage. By systematically integrating these two phases, our approach establishes diversity as the central technical design principle, enabling VibeThinker-1.5B to achieve robust performance that surpasses conventional training paradigms.
 ## Usage Guidelines

 🌱 Code Generation: It achieved scores of 55.9 on LiveCodeBench v5 and 51.1 on v6. Its v6 score slightly leads Magistral Medium (50.3), underscoring its strong reasoning performance.
+![image](https://cdn-uploads.huggingface.co/production/uploads/64d1faaa1ed6649d70d1fa2f/jYT1Iq9Jv6vw8Cllr3DuX.png)
 🔁 On the AIME 25 benchmark, VibeThinker-1.5B significantly extends the Pareto frontier of reasoning accuracy versus model scale, demonstrating that exceptional performance can be achieved with extreme parameter efficiency.
 ## Training Pipeline
+![image](https://cdn-uploads.huggingface.co/production/uploads/64d1faaa1ed6649d70d1fa2f/rPfb1GKFQUOICcFs95Aus.png)
 VibeThinker-1.5B's core innovation lies in the "Spectrum-to-Signal Principle" (SSP) training framework: it first explores solution diversity during the Supervised Fine-Tuning (SFT) stage, and then optimizes its policy to reinforce correct signals in the Reinforcement Learning (RL) stage. By systematically integrating these two phases, our approach establishes diversity as the central technical design principle, enabling VibeThinker-1.5B to achieve robust performance that surpasses conventional training paradigms.
 ## Usage Guidelines