Instructions to use cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100 with Transformers:

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

pipe = pipeline("text-generation", model="cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100")
model = AutoModelForCausalLM.from_pretrained("cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100", 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 cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100

SGLang

How to use cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100 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 "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100" \
    --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": "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100",
		"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 "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100" \
        --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": "cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100 with Docker Model Runner:
```
docker model run hf.co/cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100
```

Qwen3-1.7B — Direct-OPD (SFT-induced teacher shift), 100 steps

Pilot artifact of the Direct-OPD SFT-vs-RL policy-shift experiment (condition sft). Student Qwen/Qwen3-1.7B trained for 100 Direct-OPD steps against the token-level policy-shift signal between:

role	model
pi_T (post-shift teacher)	`cmpatino/DeepSeek-R1-Distill-Qwen-1.5B-DeepMath-SFT100` @ `baee02cc3858ddb200601f247b6a22f072b599d1`
pi_Tref (pre-shift teacher)	`deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B` @ `ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562`

Root = step 100. checkpoint-{20,40,60,80,100}/ = intermediate merged checkpoints. Weights are bf16 (verl's FSDP->HF merge downcasts the fp32 masters).

Reproduction

code: https://github.com/BytedTsinghua-SIA/Direct-OPD @ 3a9d6bd37b00a38e7a9b2959239e4631e5324aea + logs/phase4_seed.patch (seed 42 shim)
data: cmpatino/direct-opd-sft-deepmath-pilot-data @ 22625ae5db434947195bf862c429cd94504a4809 :: opd_train.parquet (6,400 prompts, one pass)
100 steps x 64 prompts x 4 rollouts, lr 1e-6, adaptive KL (init/max 2.5), max prompt 1024 / response 2048 tokens, seed 42
driver + full env block: logs/run_manifest.json, console log logs/train.log.gz

Caveats: no in-training validation (test_freq=-1); bit-exact reproducibility is not attainable (vLLM continuous batching, dynamic micro-batching, FSDP reductions); teacher scoring reuses the student's Qwen3 token ids verbatim (shared-vocab assumption, unasserted upstream).

Downloads last month: 1

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for cmpatino/Qwen3-1.7B-DirectOPD-SFTShift-100

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Finetuned

(975)

this model