Docs
Your first request
Anything that speaks the OpenAI protocol works: the Python and Node SDKs, LangChain, LlamaIndex and most desktop clients.
1. Create a key
Sign in with Google and create a key in the console.
2. Point your client at the endpoint
Change the base URL. Keep the rest of your code as it is.
3. Send a request
Streaming, tool calls and images behave the way your client expects.
Endpoint
One base URL for every model.
https://switchllm.net/v1
curl https://switchllm.net/v1/chat/completions \
-H "Authorization: Bearer $SWITCHLLM_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-flash",
"messages": [{"role": "user", "content": "Hello"}]
}'
# Streaming works the same way: add "stream": true.
Pass your key as the bearer token, and keep keys out of client-side code.
Models
Availability changes as new models land. The console shows the live list.
| Model ID | Family | Access |
|---|---|---|
| deepseek-chat | DeepSeek | Paid |
| deepseek-flash | DeepSeek | Paid |
| deepseek-reasoner | DeepSeek | Paid |
| deepseek-v4-pro | DeepSeek | Paid |
| glm-4.5 | GLM | Paid |
| glm-4.5-air | GLM | Paid |
| glm-4.5-flash | GLM | Free |
| glm-4.6 | GLM | Paid |
| glm-4.7 | GLM | Paid |
| glm-4.7-flash | GLM | Free |
| glm-5 | GLM | Paid |
| glm-5-turbo | GLM | Paid |
| glm-5.1 | GLM | Paid |
| glm-5.2 | GLM | Paid |
| glm-5.3 | GLM | Paid |
| glm-5.3-flash | GLM | Paid |
| glm-5.3-flashx | GLM | Paid |
| kimi-k2.6 | Kimi | Paid · Coming soon |
| kimi-k2.7-code | Kimi | Paid · Coming soon |