---
title: Api2
emoji: 🔥
colorFrom: indigo
colorTo: red
sdk: gradio
sdk_version: "6.20.0"
python_version: "3.12"
app_file: app.py
pinned: false
license: mit
---

# Api2 ZeroGPU Gradio Runtime

ZeroGPU edition. This Space uses Gradio SDK and `@spaces.GPU` instead of Docker/FastAPI.

Important:
- ZeroGPU only works with Gradio SDK, so Docker/vLLM/llama.cpp GGUF CUDA is not used in this variant.
- Use Transformers-format models. GGUF-only repos are listed as unsupported in the UI.
- For large models, the first request may take a long time because the model must be downloaded and loaded.