batched inference #214
Draft
batched inference #214
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
|
Does this work as expected? If you look at mlx_parallm, they use a BatchedKVCache implementation to handle the kv cache for batches https://github.com/willccbb/mlx_parallm/blob/80b18ab49b80e6f8d82d89347ab32f44b35f8942/mlx_parallm/utils.py#L201 I'm not sure how it would work with the current implementation. It looks like one cache is used for all the requests which is probably not want we want here. |
|
@AlexCheema https://github.com/willccbb/mlx_parallm/blob/80b18ab49b80e6f8d82d89347ab32f44b35f8942/mlx_parallm/utils.py#L201 is exactly the same as what I did, it's just that it stores batch_size as a variable, i just infer it from the input shape |
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Outdated suggestions cannot be applied.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.
Original PR: #108
It was built on top of older code, needs a lot of refactoringrefactored nowcurrently on pause - need to think broadly for distributed training on exo