Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
proxy: continue requests when the model catalog changes (#18058)
d366f48 Parth Sareen committed 2 months ago
launch: respect OLLAMA_HOST when configuring Pi (#18039)
91cf995 Daniel Hiltgen committed 2 months ago
mlxrunner: avoid Metal GPU timeouts when loading models from slow storage
77e3b0a Jesse Gross committed 2 months ago
mlx: return exact types and evaluate arrays in the value readers
7623501 Jesse Gross committed 3 months ago
proxy: preserve string content during image fallback (#18002)
ebf200f Parth Sareen committed 2 months ago
app: reject browser origins on Claude Desktop gateway (#17989)
6e19e91 Parth Sareen committed 2 months ago