Skip to content

Conversation

johnmai-dev
Copy link
Contributor

@johnmai-dev johnmai-dev commented Jun 21, 2025

This PR ports Falcon H1.

Thanks to my idol @Blaizzy !
Source: ml-explore/mlx-lm#231

image

@johnmai-dev johnmai-dev marked this pull request as ready for review June 22, 2025 05:14

// MARK: - Mamba2Cache KVCache

private class Mamba2Cache: KVCache {
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There is now a:

/// Simple cache for Mamba-style state space models
public class MambaCache: BaseKVCache {

added in #334 -- will that work for this? If not, would it make sense to move Mamba2Cache into KVCache.swift?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No problem, I need to take some time to study #334.

@johnmai-dev johnmai-dev marked this pull request as draft July 14, 2025 03:29
@johnmai-dev johnmai-dev changed the title Add Falcon H1 model Add Falcon H1 Jul 16, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants