AMD Radeon GPUs in the cloud

Ship on Radeon GPUs, without owning one.

A GPU instance in your browser, and an OpenAI-compatible endpoint you can curl. Both in under five minutes.

your first call

$ curl https://developer.amd.com.cn/radeon/api/v1/chat/completions \
  -H "Authorization: Bearer rc-••••" \
  -H "Content-Type: application/json" \
  -d '{"model":"DeepSeek-V4-Flash",
      "messages":[{"role":"user","content":"Hello"}]}'

{
  "object": "chat.completion",
  "model": "DeepSeek-V4-Flash",
  "choices": [{ "message": {
    "role": "assistant",
    "content": "Hello! How can I help you today?"
  } }]
}

Development

Launch a GPU instance

Create a template, launch it, and get into a ROCm-ready Radeon GPU machine through JupyterLab or SSH.

API

Call a model API

Use the free shared endpoints, or deploy a dedicated one on your own instance. Both speak the OpenAI protocol.