Skip to content

Pull requests: ml-explore/mlx-lm

New pull request New
383 Open 724 Closed
383 Open 724 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

laguna: load the public checkpoint layouts (original poolside + repacks)
#1704 opened Aug 10, 2026 by pierre427 Loading…
fix(gemma3n): share KV independently of the cache (#1674)
#1702 opened Aug 10, 2026 by Anai-Guo Loading…
1
Copy on ArraysCache.extract instead of returning a view extract slice…
#1701 opened Aug 10, 2026 by jzdziarski Loading…
Honor partial_rotary_factor in glm
#1698 opened Aug 9, 2026 by yentur Loading…
Exclude trailing padding token from the fine-tuning loss
#1697 opened Aug 9, 2026 by yentur Loading…
fix(server): fail closed on unregistered model IDs
#1696 opened Aug 9, 2026 by SashaMIT Loading…
1 task done
Require opt-in for sharded tokenizer code
#1686 opened Aug 9, 2026 by Thump604 • Draft
1
Return accurate generation error status
#1685 opened Aug 9, 2026 by Thump604 • Draft
1
Require authentication for remote server binding
#1687 opened Aug 9, 2026 by Thump604 • Draft
1
Fix perplexity all-samples mode
#1682 opened Aug 9, 2026 by Thump604 • Draft
1
Remove obsolete OLMo process exit
#1683 opened Aug 9, 2026 by Thump604 • Draft
1
Fix Python and runtime dependency metadata
#1684 opened Aug 9, 2026 by Thump604 • Draft
1
Fix uneven pipeline layer partitioning
#1680 opened Aug 9, 2026 by Thump604 • Draft
1
Validate generation batch limits
#1681 opened Aug 9, 2026 by Thump604 • Draft
1
Fix offset
#1678 opened Aug 8, 2026 by JasonHonKL Loading…
1
Fix ChunkedKVCache dropping valid tokens during decode (llama4)
#1673 opened Aug 7, 2026 by binaydhakal Loading…
Raise clear error when LoRA config uses 'alpha' instead of 'scale'
#1671 opened Aug 7, 2026 by arjunneema09 Loading…
1
Validate per-prompt max_tokens in batch_generate
#1668 opened Aug 7, 2026 by kupilikula Loading…
Add prompt checkpoints for hybrid caches
#1667 opened Aug 6, 2026 by stale2000 Loading…
Save the best-validation-loss adapter, and add optional early stopping
#1665 opened Aug 5, 2026 by axiom-of-choice Contributor Loading…
1
Fix evaluate crash on tasks with no stop sequences
#1664 opened Aug 5, 2026 by choipilkyu Loading…
1
Add Apertus tool parser
#1663 opened Aug 5, 2026 by gabriben Loading…
Add optional gradient clipping and non-finite guard to the LoRA trainer
#1659 opened Aug 5, 2026 by axiom-of-choice Contributor Loading…
1
2
fix(generate): guard per-sequence logits_processors against None in batched generation
#1657 opened Aug 4, 2026 by chronoshift Loading…
Add support for Kanana-2 Tiny (kanana2_tiny)
#1655 opened Aug 4, 2026 by choipilkyu Loading…
ProTip! Exclude everything labeled bug with -label:bug.