Chat
Messages, images and documents go to the selected model.
Test your own model
Enter your server's address and key on the right, press Connect and pick a model.
Works with vLLM, llama.cpp, Ollama, LM Studio, TGI, LocalAI, LiteLLM and anything else that speaks the OpenAI API.
Connection
Your server and key stay in this browser.
Connection route and settings
Route
Health
Live numbers from every reply.
Not connected yet
Connect to a server to see how it's doing.
Generation speed
–
First token
–
Prompt reading
–
tokens per second
Response time
–
model list
Last reply
–
tokens in · tokens out · total time