Skip to content

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

Commits on main

mlx: speed up Qwen 3.8 prompt processing (#18550)
Daniel Hiltgen committed 1 month ago
c0f8da3
b8c3d1f
MLX: version bump (#18576)
Daniel Hiltgen committed 1 month ago
421d74d
llama.cpp: version update (#18577)
Daniel Hiltgen committed 1 month ago
13e0247
a9d8953
d0c8cdb
tests: fix metadata deletion test flake (#18493)
Daniel Hiltgen committed 2 months ago
4f6f739
nemotron_h: add MLX vision support (#17714)
Daniel Hiltgen committed 2 months ago
2c73164
ci: harden release artifact uploads (#18516)
Daniel Hiltgen committed 2 months ago
5113242
ec3cc23
app: add Apps deep link (#18497)
Eva H committed 2 months ago
5ed8dde
6b8bef2
28a9f8c
mlxrunner: lay out nn one layer kind per file
Jesse Gross committed 2 months ago
7343e22
d27fde6
mlx, mlxrunner: move the MLX engine out of x/
Jesse Gross committed 2 months ago
2e036e7
tokenizer, ml: remove dead code
Jesse Gross committed 2 months ago
bef41f7
6a0ee48