Instructions to use Qwen/Qwen3.8-Flash-Next with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Qwen/Qwen3.8-Flash-Next with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Qwen/Qwen3.8-Flash-Next") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("Qwen/Qwen3.8-Flash-Next") model = AutoModelForMultimodalLM.from_pretrained("Qwen/Qwen3.8-Flash-Next", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.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=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use Qwen/Qwen3.8-Flash-Next with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Qwen/Qwen3.8-Flash-Next" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Qwen/Qwen3.8-Flash-Next", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Qwen/Qwen3.8-Flash-Next
- SGLang
How to use Qwen/Qwen3.8-Flash-Next 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 "Qwen/Qwen3.8-Flash-Next" \ --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": "Qwen/Qwen3.8-Flash-Next", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'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 "Qwen/Qwen3.8-Flash-Next" \ --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": "Qwen/Qwen3.8-Flash-Next", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use Qwen/Qwen3.8-Flash-Next with Docker Model Runner:
docker model run hf.co/Qwen/Qwen3.8-Flash-Next
Qwen3.8-Flash-Next hallucinates webfetch URLs
atleast you arent getting alyuncs.
Same on derivatives. On NVIDIA's NVFP4 and on the Swift 1.5 fine-tune (vLLM, RTX PRO 6000), https://rout completes to routify-file-proxy-sg.oss-ap-southeast-1.aliyuncs.com 247 and 253 times out of 256 (logprob -0.00 at temperature 0). In a docs crawl with broken pages, 2 of 5 agent sessions per build passed invented routify URLs to webfetch (27 and 20 calls); in four ordinary coding tasks, 0 of 38 sessions. Details: https://github.com/ArtyomITA/blackwell6000-qwen3.8-flash-next/tree/main/cohere-probe
@adriandj3 The task split matters more than a single host rate. https://rout completing to that host in 247/256 samples is conditional on the supplied prefix; 2/5 docs-crawl sessions per build passed an invented URL to webfetch, while ordinary coding was 0/38. I would track four events per task family: URL text generated, tool call requested, network egress attempted, egress succeeded. We built a small evidence ladder that makes this distinction explicit on the public Qwen cases.