Instructions to use RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k with Transformers:

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

pipe = pipeline("text-generation", model="RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k")
model = AutoModelForCausalLM.from_pretrained("RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k", 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 RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k

SGLang

How to use RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k 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 "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k" \
    --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": "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k",
		"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 "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k" \
        --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": "RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k with Docker Model Runner:
```
docker model run hf.co/RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k
```

Qwen3-4B distilled from Qwen3-30B-A3B — On-policy 5k (code)

On-policy KD for a Qwen3-4B student toward the Qwen3-30B-A3B teacher, on 5,000 code prompts from OpenThoughts-3.

Domain-transfer companion to qwen3_30b_a3b_to_4b_onpolicy_science_5k and the math-domain onpolicy_5k_src* chain: same student starting point and recipe, different domain slice. Training exposure of this checkpoint:

Off-policy KD on 20k math prompts → qwen3_30b_a3b_to_4b_offpolicy_20k
On-policy KD (this run) on 5k code prompts → this checkpoint

Models

Role	Model
Student	RockToken/qwen3_30b_a3b_to_4b_offpolicy_20k
Teacher	Qwen/Qwen3-30B-A3B-Instruct-2507 (MoE)

enable_thinking=False throughout.

Training data

Source: open-thoughts/OpenThoughts3-1.2M, domain == "code" slice
5,000 single-user-turn prompts; 70 exceeding prompt_max_len=2560 were filtered, leaving 4,930
Only the prompts are used; on-policy KD never reads the dataset's reference answers

Training setup

Framework: KDFlow — FSDP2 + SGLang rollout, Ray-orchestrated GPU co-location with sleep/wakeup. Same pipeline revision as the math-domain onpolicy_5k_src* runs.

Hardware: 1× node, 4× H100 (94 GB), 33 h 09 min wall-clock (~133 GPU-hours).

Key hyperparameters

Group	Value
Backend	fsdp2, bf16, gradient ckpt on
Epochs	1 (2,465 rollout iterations)
Train batch	4 (micro 1)
Learning rate	2e-6, cosine, warmup 5%
KD ratio	1.0
KD loss	reverse KL (`rkl`)
KD algorithm	`vanilla_kd`
Rollout engine	SGLang, TP=2, 1 engine
Rollout batch	2 prompts × 4 samples/prompt
`generate_max_len`	8000
`prompt_max_len`	2560 (total `max_len` 10816)
Sampling	temperature 1.0, top-p 1.0
Teacher	TP=4, sleep/wakeup enabled

prompt_max_len is 2560 instead of the math runs' 800: code prompts carry problem statements and starter code (p99 ≈ 2,356 tokens).

Note on the LR schedule: as in the earlier runs of this series, the scheduler's horizon is sized to 4× the actual number of updates, so the cosine decay is only partially traversed — the LR moves from 2.00e-6 to 1.79e-6 over the run rather than reaching zero. Kept as is for comparability within the series.

Training dynamics

Means over 500-step blocks (per-step values are noisy at batch 4):

steps	loss (reverse KL)	mean gen length
1–500	0.190	6,814
501–1000	0.178	6,611
1001–1500	0.175	6,471
1501–2000	0.166	6,647
2001–2465	0.166	6,378

The loss starts near 0.19 rather than the ~3 seen when distilling from a raw instruct model — the student has already been through off-policy KD against the same teacher, so on-policy training starts close and grinds out the remaining distribution gap.

⚠ Truncation caveat

Code responses run long, and a substantial share of rollouts hit the 8,000-token generation cap: per-step mean lengths sit at p50 = 7,069 against the cap, and in 20% of steps the whole batch was truncated (step-mean ≥ 7,990). Truncated rollouts contribute KD signal only for their first 8,000 tokens, and the student receives no supervision on how the teacher would have ended those responses. Treat comparisons against the science and math checkpoints (which trained essentially truncation-free) with this in mind.

Intended use

Research on distillation dynamics, in particular whether on-policy KD gains transfer across domains (math-trained off-policy base → code on-policy round). Domain: code (OpenThoughts-3 code split).

Limitations

The on-policy round is code-only; not tuned for chat or safety.
enable_thinking=False — this student does not emit thinking traces.
Heavy generation-length truncation during training; see the caveat above.
Single checkpoint at the end of one epoch; no intermediate checkpoints were kept.

Downloads last month: 24

Safetensors

Model size

4B params

Tensor type

F32

Model tree for RockToken/qwen3_30b_a3b_to_4b_onpolicy_code_5k

Base model

RockToken/qwen3_30b_a3b_to_4b_offpolicy_20k

Finetuned

(4)

this model

RockToken
/

qwen3_30b_a3b_to_4b_onpolicy_code_5k