Instructions to use jackcai1206/qwen3-4b-longcot-sft-vanilla with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jackcai1206/qwen3-4b-longcot-sft-vanilla with Transformers:

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

pipe = pipeline("text-generation", model="jackcai1206/qwen3-4b-longcot-sft-vanilla")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jackcai1206/qwen3-4b-longcot-sft-vanilla")
model = AutoModelForCausalLM.from_pretrained("jackcai1206/qwen3-4b-longcot-sft-vanilla", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jackcai1206/qwen3-4b-longcot-sft-vanilla with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jackcai1206/qwen3-4b-longcot-sft-vanilla"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jackcai1206/qwen3-4b-longcot-sft-vanilla",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jackcai1206/qwen3-4b-longcot-sft-vanilla

SGLang

How to use jackcai1206/qwen3-4b-longcot-sft-vanilla 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 "jackcai1206/qwen3-4b-longcot-sft-vanilla" \
    --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": "jackcai1206/qwen3-4b-longcot-sft-vanilla",
		"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 "jackcai1206/qwen3-4b-longcot-sft-vanilla" \
        --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": "jackcai1206/qwen3-4b-longcot-sft-vanilla",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jackcai1206/qwen3-4b-longcot-sft-vanilla with Docker Model Runner:
```
docker model run hf.co/jackcai1206/qwen3-4b-longcot-sft-vanilla
```

Qwen3-4B long-CoT SFT

A vanilla supervised fine-tune of Qwen/Qwen3-4B-Base on long chain-of-thought math data. No architectural changes — a standard Qwen3 model, loadable with from_pretrained.

We release it as a full-context reference baseline: a plain SFT of this base model at this data scale, useful as a control to measure other fine-tunes against, and as a frozen teacher for distillation.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

REPO = "jackcai1206/qwen3-4b-longcot-sft-vanilla"
model = AutoModelForCausalLM.from_pretrained(REPO, dtype="bfloat16", device_map="auto")
tok   = AutoTokenizer.from_pretrained(REPO)

Precision

Training kept fp32 master weights (16.1 GB). Downstream use constructs the model in bf16, so those weights are truncated to bf16 on load — this export is bf16 and therefore bit-identical to what training and inference actually consume, at half the size. Verified tensor by tensor against the fp32 checkpoint. lm_head is tied to embed_tokens in the source checkpoint (same storage), so it is reconstructed by from_pretrained rather than stored separately.

Evaluation

MATH-500 (strict): 87.4.

Downloads last month: 49

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for jackcai1206/qwen3-4b-longcot-sft-vanilla

Base model

Qwen/Qwen3-4B-Base

Finetuned

(463)

this model