Skip to content

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Commits on main

server: cache show responses (#15967)
Parth Sareen committed 6 months ago
d319227
mlx: partial cleanup of imagegen layout (#15435)
Daniel Hiltgen committed 6 months ago
2d84ec9
mlx: Gemma4 MTP speculative decoding (#15980)
Patrick Devine committed 6 months ago
15e6076
go: bump to 1.26 (#15904)
Parth Sareen committed 6 months ago
4017af9
Update MLX and MLX-C with threading fixes (#15845)
Daniel Hiltgen committed 6 months ago
534342e
launch: claude app (#15937)
Parth Sareen committed 6 months ago
9ba5a04
metal: harden for ggml initialization failures (#15755)
Daniel Hiltgen committed 6 months ago
4fe5609
917324b
renderers: update gemma4 renderer (#15886)
Parth Sareen committed 6 months ago
c7c2837
b6447ca
launch/docs: fix title for pool (#15883)
Eva H committed 6 months ago
bad32c7
New models (#15861)
Daniel Hiltgen committed 6 months ago
87288ce
2bbe240
mlxrunner: apply RoPE at per-row positions
Jesse Gross committed 7 months ago
bd21678
088dfd8
03aee88
tokenizer: fix multi-regex BPE offset handling (#15844)
Daniel Hiltgen committed 6 months ago
ec9b4e9
4656a07
30f86cb
ea01af6
api: accept "max" as a think value (#15787)
Parth Sareen committed 7 months ago
c2ebb4d
launch: harden OpenClaw onboarding flow (#15777)
Parth Sareen committed 7 months ago
590109c
160660e