*Major T/s improvement* Use the Metal qmatmul MM kernels #2615
+5,612
−2,523
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Add GGUF bf16 type support * Add non avx impl for vec_dot_bf16 * Fix from_u32 * Fix loading * Fix dequant of bf16
* Update kernels for metal bf16 * Fix typo * Check if have bfloat
* Test passes * All tests pass * Now all the tests really pass * Try out always using mm * Mirror llama.cpp metric * Mirror llama.cpp metric * Update test
|
@LaurentMazare if you could review, that would be great! More benchmarks with some smaller models can be found here: EricLBuehler/mistral.rs#903 (comment) |
This was referenced Nov 22, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds the automatic usage of Metal GGML quantized mat-mat kernels instead of always using the mat-vec kernels and upstreams a few related/necessary changes.
Before this change, Candle's Metal decoding performance was on-par with MLX and llama.cpp but the prompt performance was insufficient. After this change, the prompt performance (on the benchmark) was increased to a factor of about 2.5x faster than MLX and within 10% of llama.cpp - a performance boost by a factor of almost 6x.
This PR switches to only using the MV kernels when
D::Minus2of thexsinput tensor is equal to 1. This mirrors the logic in GGML.Besides utilizing the MM kernels, this PR also upstreams some required changes: