AlayaCode overview
Subscription AI coding assistant — Chinese LLMs aggregated under OpenAI / Anthropic dual-compatible APIs
AlayaCode is Alaya NeW Cloud's subscription AI coding assistant for developers:
- Aggregates Chinese LLMs: MiniMax, Zhipu GLM, DeepSeek, and more behind one endpoint
- Dual protocol: the same API key works for OpenAI Chat Completions and Anthropic Messages
- Subscription billing: fixed monthly quota, no surprise bills; throttled (not cut off) when quota exhausts
- Out-of-the-box IDE integration: Cursor, Continue, Cline, Roo, Claude Code, Codex CLI and friends connect directly
Three steps to first call
One-time setup
Sign up at www.alayanew.com/backend/register, complete identity verification, then go to AlayaCode → API Key to mint a subscription key. The free trial credit (¥30) covers every example on this page.
1. Pick a protocol
| Protocol | Base URL | Works with |
|---|---|---|
| OpenAI | https://codingplan.alayanew.com/v1 | Cursor, Continue, OpenAI SDK, Codex CLI, native openai Python / Node libs |
| Anthropic | https://codingplan.alayanew.com/anthropic | Claude Code, Cline, Roo Code, native anthropic SDK |
Both protocols hit the same underlying models — pick whichever your tool prefers; you don't burn quota on the "wrong" choice.
2. Set environment variables
export OPENAI_BASE_URL=https://codingplan.alayanew.com/v1
export OPENAI_API_KEY=sk-your-key-hereexport ANTHROPIC_BASE_URL=https://codingplan.alayanew.com/anthropic
export ANTHROPIC_AUTH_TOKEN=sk-your-key-here3. Make your first call
curl https://codingplan.alayanew.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer sk-your-key-here" \
-d '{
"model": "minimax-m2.5",
"messages": [
{"role": "user", "content": "Hello"}
]
}'curl https://codingplan.alayanew.com/anthropic/v1/messages \
-H "Content-Type: application/json" \
-H "x-api-key: sk-your-key-here" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "minimax-m2.5",
"max_tokens": 1024,
"messages": [
{"role": "user", "content": "Hello"}
]
}'Where to go next
API reference
Endpoints, request bodies, responses
SDK usage
Reuse the official OpenAI / Anthropic SDKs directly
IDE & tools
Cursor, Continue, Claude Code, Cline, Codex, …
Assistant commands
/help, /usage, /reset and other slash commands
Errors
HTTP and business error codes
Billing & plans
Subscription tiers, quota and rate limits
Supported models
| Model ID | Provider | Context | Best for |
|---|---|---|---|
minimax-m2.5 | MiniMax | 200K | Flagship for general chat + code generation, fastest path |
minimax-m2.1 | MiniMax | 200K | Previous stable, cheaper |
glm-5 | Zhipu | 200K | Strongest Chinese, very stable long-doc summarisation |
glm-5.1 | Zhipu | 200K | Code-tuned GLM-5; better function-calling quality |
deepseek-v4-pro | DeepSeek | 1M | Long-context reasoning workloads |
The available list shifts with your subscription tier; AlayaCode → Usage in the console always shows the live set.
Last updated on
