Instructions to use z-lab/Qwen3.5-9B-DFlash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use z-lab/Qwen3.5-9B-DFlash with Transformers:

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

pipe = pipeline("text-generation", model="z-lab/Qwen3.5-9B-DFlash", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("z-lab/Qwen3.5-9B-DFlash", trust_remote_code=True)
model = AutoModel.from_pretrained("z-lab/Qwen3.5-9B-DFlash", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use z-lab/Qwen3.5-9B-DFlash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "z-lab/Qwen3.5-9B-DFlash"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "z-lab/Qwen3.5-9B-DFlash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/z-lab/Qwen3.5-9B-DFlash

SGLang

How to use z-lab/Qwen3.5-9B-DFlash 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 "z-lab/Qwen3.5-9B-DFlash" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "z-lab/Qwen3.5-9B-DFlash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "z-lab/Qwen3.5-9B-DFlash" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "z-lab/Qwen3.5-9B-DFlash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use z-lab/Qwen3.5-9B-DFlash with Docker Model Runner:
```
docker model run hf.co/z-lab/Qwen3.5-9B-DFlash
```

Qwen3.5-9B-DFlash / README.md

jianchen0311

Update README.md

9b697fb verified 6 months ago

preview code

Raw

History Blame

2.13 kB

	---
	license: apache-2.0
	library_name: transformers
	pipeline_tag: text-generation
	tags:
	- speculative-decoding
	- diffusion
	- efficiency
	- flash-decoding
	- qwen
	- diffusion-language-model
	---

	# Qwen3.5-9B-DFlash
	[Paper](https://arxiv.org/abs/2602.06036) \| [GitHub](https://github.com/z-lab/dflash) \| [Blog](https://z-lab.ai/projects/dflash/)

	This model is still under training.

	DFlash is a novel speculative decoding method that utilizes a lightweight block diffusion model for drafting. It enables efficient, high-quality parallel drafting that pushes the limits of inference speed.

	This model is the drafter component. It must be used in conjunction with the target model `Qwen/Qwen3.5-9B`. It was trained with a context length of 4096 tokens.

	<div align="center">
	<img src="assets/dflash_system.png" alt="DFlash Architecture" width="100%">
	</div>

	## 🚀 Quick Start

	### SGLang

	#### Installation
	```bash
	uv pip install "git+https://github.com/sgl-project/sglang.git@refs/pull/16818/head#subdirectory=python"
	```

	#### Inference
	```bash
	python -m sglang.launch_server \
	--model-path Qwen/Qwen3.5-9B \
	--speculative-algorithm DFLASH \
	--speculative-draft-model-path z-lab/Qwen3.5-9B-DFlash \
	--speculative-num-draft-tokens 16 \
	--tp-size 1 \
	--dtype bfloat16 \
	--attention-backend fa3 \
	--mem-fraction-static 0.75 \
	--trust-remote-code \
	--mamba-scheduler-strategy extra_buffer \
	--reasoning-parser qwen3 \
	--tool-call-parser qwen3_coder
	```
	> Note: For long-context or agentic usage (such as OpenClaw or Claude Code), consider adding `--speculative-dflash-draft-window-size WINDOW_SIZE` to enable sliding-window attention for the draft model. Because the draft model is only trained on 4K context, this often improves performance on very long context (50K+ tokens).

	#### Early Results
	- Thinking: enabled
	- Max new tokens: 4096
	- Block size: 16
	\| Dataset \| Accept Length \|
	\|-----------\|---------------\|
	\| GSM8K \| 6.709 \|
	\| Math500 \| 7.388 \|
	\| HumanEval \| 7.888 \|
	\| MBPP \| 6.617 \|
	\| MT-Bench \| 5.506 \|
	\| Alpaca \| 5.079 \|