Instructions to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT with PEFT:
```
Task type is invalid.
```

How to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT with Transformers:

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

pipe = pipeline("text-generation", model="RationalPursuit/Qwen3-4B-Stratos-R1-SFT")
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("RationalPursuit/Qwen3-4B-Stratos-R1-SFT")
model = AutoModelForCausalLM.from_pretrained("RationalPursuit/Qwen3-4B-Stratos-R1-SFT", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RationalPursuit/Qwen3-4B-Stratos-R1-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RationalPursuit/Qwen3-4B-Stratos-R1-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RationalPursuit/Qwen3-4B-Stratos-R1-SFT

SGLang

How to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT 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 "RationalPursuit/Qwen3-4B-Stratos-R1-SFT" \
    --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": "RationalPursuit/Qwen3-4B-Stratos-R1-SFT",
		"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 "RationalPursuit/Qwen3-4B-Stratos-R1-SFT" \
        --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": "RationalPursuit/Qwen3-4B-Stratos-R1-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RationalPursuit/Qwen3-4B-Stratos-R1-SFT with Docker Model Runner:
```
docker model run hf.co/RationalPursuit/Qwen3-4B-Stratos-R1-SFT
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Model Card

Experimental research artifact. Base: Qwen/Qwen3-4B-Instruct-2507. SFT on bespokelabs/Bespoke-Stratos-17k DeepSeek-R1 traces (14,815 paired rows). Named -SFT, not -Distill — we SFT'd on teacher traces; we did not run the distillation ourselves.

Eval (Stratos held-out, 508 prompts): non-termination 109/508 = 0.215; has_think_block 399/508.

The collection shares a base and recorded LoRA settings. This student retained 14,815 training questions; sampled Qwen retained 14,772 and gpt-oss retained 14,768. Teacher generation and curation conditions also differ, so comparisons do not isolate teacher identity as the sole cause of behavior differences. See the source records.