Instructions to use meituan-longcat/LongCat-Flash-Chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meituan-longcat/LongCat-Flash-Chat with Transformers:

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

pipe = pipeline("text-generation", model="meituan-longcat/LongCat-Flash-Chat", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("meituan-longcat/LongCat-Flash-Chat", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use meituan-longcat/LongCat-Flash-Chat with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/meituan-longcat/LongCat-Flash-Chat

SGLang

How to use meituan-longcat/LongCat-Flash-Chat 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 "meituan-longcat/LongCat-Flash-Chat" \
    --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": "meituan-longcat/LongCat-Flash-Chat",
		"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 "meituan-longcat/LongCat-Flash-Chat" \
        --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": "meituan-longcat/LongCat-Flash-Chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use meituan-longcat/LongCat-Flash-Chat with Docker Model Runner:
```
docker model run hf.co/meituan-longcat/LongCat-Flash-Chat
```

inconsistency in the `configuration_longcat_flash.py`

#11

by hb4ch - opened Sep 23, 2025

Discussion

hb4ch

Sep 23, 2025

in the comment section:
norm_topk_prob (`bool`, *optional*, defaults to `True`): Whether to normalize the weights of the routed experts.
Whereas in the function signature: def __init__( self, ... norm_topk_prob=False, ... )
This could cause confusion and lead to incorrect community implementation.

I suggest that either you could add this param explicitly in the configuration file, or keep configuration_longcat_flash.py consistent

LongCat0830

LongCat org Sep 24, 2025

Thanks for your reminder, this file has been updated. @hb4ch

LongCat0830 changed discussion status to closed Sep 24, 2025

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