Spaces:
Running on Zero
Running on Zero
A newer version of the Gradio SDK is available: 6.20.0
metadata
title: VibeThinker 3B ZeroGPU Chat
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 6.18.0
python_version: 3.12
app_file: app.py
startup_duration_timeout: 1h
hardware: zero-a10g
pinned: false
license: mit
models:
- WeiboAI/VibeThinker-3B
VibeThinker-3B ZeroGPU Chat
This Space hosts WeiboAI/VibeThinker-3B on Hugging Face ZeroGPU.
The frontend is the upstream huggingface/chat-ui SvelteKit app vendored in chat-ui/ at commit e1b4806068a0312752d9ee0acc48bb8d9ae24b71.
app.py runs Gradio Server mode for the ZeroGPU-backed model API, exposes local OpenAI-compatible /v1/models and /v1/chat/completions routes, starts chat-ui on an internal Node server, and reverse-proxies browser traffic to chat-ui.