Instructions to use z-lab/Qwen3.6-27B-DFlash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use z-lab/Qwen3.6-27B-DFlash with Transformers:

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

pipe = pipeline("text-generation", model="z-lab/Qwen3.6-27B-DFlash", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("z-lab/Qwen3.6-27B-DFlash", trust_remote_code=True)
model = AutoModel.from_pretrained("z-lab/Qwen3.6-27B-DFlash", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use z-lab/Qwen3.6-27B-DFlash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "z-lab/Qwen3.6-27B-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/Qwen3.6-27B-DFlash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/z-lab/Qwen3.6-27B-DFlash

SGLang

How to use z-lab/Qwen3.6-27B-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/Qwen3.6-27B-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/Qwen3.6-27B-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/Qwen3.6-27B-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/Qwen3.6-27B-DFlash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use z-lab/Qwen3.6-27B-DFlash with Docker Model Runner:
```
docker model run hf.co/z-lab/Qwen3.6-27B-DFlash
```

Why is there a 4k ctx limit?

by crazyi - opened 10 days ago

Discussion

crazyi

10 days ago

I am using omlx, but the maximum size of the ctx is only 4096. I don’t know why there is this limitation. Is it due to omlx or dflash?
https://github.com/jundot/omlx/blob/main/docs/experimental/dflash_mlx_integration.md

jianchen0311

Z Lab org 10 days ago

It shouldn't. The latest DFlash draft model with SWA layers has been able to support much longer context length ( In my test it still works with 100K context length when using Claude Code). So this is omlx issue, they should remove this limit.

miabchdave

10 days ago

•

edited 10 days ago

You can start the regular oMLX application in a terminal with whatever context that you want for DFlash by using its environment variable. For example, for contexts up to 8192, use the following: DFLASH_MAX_CTX=8193 /Applications/oMLX.app/Contents/MacOS/omlx-cli serve

I think they're soon adding an option for the DFlash context in the UI.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment