fara-7b-api / README.md
likhonhfai's picture
Upload folder using huggingface_hub
cb655e6 verified
metadata
title: Fara 7B API
emoji: 🤖
colorFrom: blue
colorTo: indigo
sdk: docker
pinned: false
license: mit

Fara-7B API Deployment

This Space hosts the Qwen/Qwen3-VL-8B-Thinking model using SGLang and provides custom API endpoints.

API Endpoints

  • GET /: Root info
  • GET /health: Health check
  • POST /v1/responses: Simple text generation
  • POST /v1/messages: Chat-style generation

Hardware Requirements

This model (7B parameters) requires a GPU for efficient inference. Please ensure the Space is running on a GPU-enabled instance (e.g., T4, L4, or A10G).