Skip to content

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

Commits on main

server: remove duplicate template parsing (#16287)
Anto jones committed 6 months ago
632ff00
275f122
create: read draft architecture from its config.json
Jesse Gross committed 6 months ago
3256853
mlxrunner: move YaRN RoPE helpers into x/models/nn
Jesse Gross committed 6 months ago
438fb99
358af4a
launch: enriched model inventory (#16230)
Parth Sareen committed 6 months ago
91c8e5e
Reduce startup model hydration (#16215)
Daniel Hiltgen committed 6 months ago
4b2d529
codex: omit patch tool type (#16231)
Bruce MacDonald committed 6 months ago
e6b1d75
56b319f
ci: speed up release builds (#15982)
Daniel Hiltgen committed 6 months ago
42e6f56
quiet down kv log spew (#16105)
Daniel Hiltgen committed 6 months ago
da679ad
docs: add codex app (#16163)
Parth Sareen committed 6 months ago
b9c0421
mlxrunner: add DFlash speculative decoding (#16134)
Patrick Devine committed 6 months ago
98e26b8
launch: codex app restarts (#16155)
Parth Sareen committed 6 months ago
c28ddc0
launch: update codex app UI copy (#16157)
Parth Sareen committed 6 months ago
3ad2fa3
docs: hide codex app till launch (#16153)
Parth Sareen committed 6 months ago
6b6f45e
mlx: rework the MLX sampler (#16122)
Patrick Devine committed 6 months ago
4860130
launch: codex app integration (#16120)
Parth Sareen committed 6 months ago
ac7295c
mlx: add memory trace logging (#16131)
Daniel Hiltgen committed 6 months ago
6398cd5
mlx: fix macOS 26 target leakage in v3 metallib (#16053)
Daniel Hiltgen committed 6 months ago
421faa0
mlx: avoid status timeout during inference (#16086)
Daniel Hiltgen committed 6 months ago
206b049
d819ef0
app: harden update flows (#16100)
Daniel Hiltgen committed 6 months ago
3d5a011
test: integration test hardening (#13532)
Daniel Hiltgen committed 6 months ago
c2f2d90
mlx: refined model push behavior (#15431)
Daniel Hiltgen committed 6 months ago
1e1b34d
launch: disable Claude Desktop launch (#16028)
Parth Sareen committed 6 months ago
f866e76
launch: add plan-aware model gating (#16027)
Parth Sareen committed 6 months ago
bab5907
7c2c36b