Daniel Hiltgen c2b0bb7a52 mlx: update as of 3/23 (#14789) ...

* mlx: update to HEAD on 3/23

Also fixes a few misc vendoring bugs uncovered with this first update.
This also renames the version files to make them clearer.

* CUDA Fast Gated Delta kernel

* mlx: detect eval errors and panic

On model errors or missing kernels, don't mask the error, bubble it up.

2026-03-23 11:28:44 -07:00

..

cache

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

mlx

mlx: update as of 3/23 (#14789)

2026-03-23 11:28:44 -07:00

model

mlx: quantized embeddings, fast SwiGLU, and runtime fixes (#14884)

2026-03-17 11:21:38 -07:00

sample

mlxrunner: fix Slice(0, 0) returning full dimension instead of empty

2026-03-18 16:06:33 -07:00

cache_test.go

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

cache_trie_test.go

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

cache_trie.go

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

cache.go

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

client.go

mlx: update as of 3/23 (#14789)

2026-03-23 11:28:44 -07:00

imports.go

MLX: add header vendoring and remove go build tag (#14642)

2026-03-09 17:24:45 -07:00

pipeline.go

mlxrunner: share KV cache across conversations with common prefixes

2026-03-18 16:06:33 -07:00

runner.go

MLX: add header vendoring and remove go build tag (#14642)

2026-03-09 17:24:45 -07:00

server.go

mlx: quantized embeddings, fast SwiGLU, and runtime fixes (#14884)

2026-03-17 11:21:38 -07:00

utf8_buffer_test.go

consolidate the tokenizer (#14327)

2026-02-19 15:55:45 -08:00

utf8_buffer.go

consolidate the tokenizer (#14327)

2026-02-19 15:55:45 -08:00