AMD Radeon Cloud Free APIs
- Expiry date
- No fixed end date
- Updated
- Jan 1, 2026
AMD Radeon Cloud's Token Factory provides developers with ready-to-use public free API endpoints powered by AMD GPU Cloud infrastructure, eliminating the need to provision dedicated hardware for evaluation and prototyping.
Available Free Endpoints
- DeepSeek-V4-Flash-Vision-Exp: Vision-language model (VLM) endpoint hosted on AMD GPU Cloud.
- DeepSeek-V4-Flash-0731: High-throughput text LLM endpoint for general generation and reasoning.
- Qwen3.8-Flash-Next: Multimodal VLM endpoint served directly via Radeon Cloud.
- MiniCPM5-2B: Lightweight text model from OpenBMB provided with limited free API access.
Platform Features
- Ready-to-Use APIs: Direct endpoint access without requiring dedicated server deployment or credit consumption.
- OpenAI-Compatible Interfaces: Easily plug models into existing OpenAI client libraries and frameworks.
- AMD GPU Acceleration: Inference powered and optimized on AMD Radeon hardware.