Use GLM, Qwen and Kimi with one API.
FreeLLMAPI gives overseas developers one endpoint for leading Chinese AI models. Use OpenAI Chat Completions or Anthropic Messages with the same FreeLLMAPI key.
OpenAI format
curl https://api.freellmapi.ai/v1/chat/completions \
-H "Authorization: Bearer sk_live_xxx" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"messages": [
{"role": "user", "content": "Write a Python function"}
]
}'
Anthropic Messages format
curl https://api.freellmapi.ai/v1/messages \
-H "x-api-key: sk_live_xxx" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"max_tokens": 512,
"messages": [
{"role": "user", "content": "Write a Python function"}
]
}'
OpenAI SDK compatible
Works with common OpenAI-compatible clients by changing the base URL and model name.
Anthropic Messages API
Works with Claude-style clients that can point to a custom Anthropic-compatible base URL.
10,000 free tokens
New verified accounts receive starter tokens automatically for quick testing.
China model coverage
FreeLLMAPI makes leading Chinese AI models easier to use from overseas. GLM, Qwen 3.8 preview, and Kimi K3 are live now through one compatible API.
Z.ai / GLM
Current gateway models include glm-5.2, glm-5.1, and glm-4.7. glm-5.2 is the latest verified route; glm-4.7 remains available as a lower-load fallback.
Alibaba / Qwen
qwen3.8-max-preview is live for advanced reasoning, coding, and agent workflows.
DeepSeek
DeepSeek V3.2 and V4-family models are tracked for reasoning and low-cost coding use cases.
Moonshot / Kimi
kimi-k3 is live for coding, reasoning, and agentic workflows through OpenAI and Anthropic formats.
Choose from Kimi K3, Qwen 3.8 Max Preview, and three verified GLM routes.
/v1/chat/completions for OpenAI format and /v1/messages for Anthropic Messages format.