Instructions to use StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35 with Transformers:

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

pipe = pipeline("text-generation", model="StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35")
model = AutoModelForCausalLM.from_pretrained("StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35", 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 StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35

SGLang

How to use StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35 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 "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35" \
    --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": "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35",
		"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 "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35" \
        --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": "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35 with Docker Model Runner:
```
docker model run hf.co/StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35
```

Qwen3.5-4B SWE-OPD baseline-simple (step 35)

On-policy distillation (OPD) checkpoint of Qwen3.5-4B (text) after 1 epoch / 35 steps on the tmax15k Harbor SWE set.

This is the merged Hugging Face export of the FSDP actor checkpoint:

swe_opd_baseline_simple / baseline_simple_1ep_bs12_lr5e7 / global_step_35

Training

Item	Value
Base model	Qwen3.5-4B text (`Qwen3_5ForCausalLM`)
Method	On-policy distillation (OPD)
Dataset	tmax15k_top500_harbor_index30
Steps	35 (1 epoch)
Train batch size	12
Actor LR	5e-7
Max response length	8192
Distillation top-k	8
Student parallel	12 GPU FSDP, DP=12
Teacher	tmax-9b
Teacher probe / eval-turn mask	off (baseline-simple)

Files

Merged from 12 FSDP shards (model_world_size_12_rank_*.pt) to a single bfloat16 model.safetensors.

Usage

Requires transformers>=5.6.0.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, dtype="bfloat16", device_map="auto")

Downloads last month: -

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for StephYang/qwen3.5-4b-swe-opd-baseline-simple-step35

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(606)

this model