Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
Commits on main
model/renderers: match Muse Glimmer reasoning template (#17732)
7ce88bd Daniel Hiltgen committed 3 months ago
mlx: avoid pulling MLX models when MLX is missing (#17710)
8831349 Daniel Hiltgen committed 3 months ago
server/images: prevent skipVerify map collision with duplicate digests (#15504)
4138e85 Vignesh committed 3 months ago
api: stop applying repeat_penalty 1.1 to models that don't set one
6a261db Jesse Gross committed 3 months ago
nemotron_h: support the Nemotron 3.5 prompt layout (#17672)
96fb6d2 Daniel Hiltgen committed 3 months ago
parsers: recover boundary tokens fumbled into glimmer ATEM invoke names (#17664)
400164d Daniel Hiltgen committed 3 months ago
mlxrunner: feed media features to the model during prefill
5fcf71b Jesse Gross committed 3 months ago
create: keep vision towers at source precision when quantizing
8713570 Jesse Gross committed 3 months ago
manifests: remove OCI rootfs from the model config (#17619)
5a173ed Daniel Hiltgen committed 3 months ago