Skip to content

Pull requests: ml-explore/mlx-lm

New pull request New
51 Open 540 Closed
51 Open 540 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

Bump the patch version
#959 opened Mar 7, 2026 by angeloskath Loading…
Reformat models for the rope parameter args
#953 opened Mar 6, 2026 by Goekdeniz-Guelmez Loading…
2
Improve OOM handling and add memory controls for long-running coding agent sessions
#948 opened Mar 4, 2026 by dmitryryabkov Loading…
Add --served-model-name flag to server
#947 opened Mar 3, 2026 by otarkhan Loading…
2 tasks done
1
Support KV cache quantization with continuous batching
#941 opened Mar 2, 2026 by ochafik Loading…
3 tasks done
1
feat: add --kv-bits CLI args for server KV cache quantization
#934 opened Feb 27, 2026 by lichengzhe Loading…
feat: add context_length to /v1/models response
#933 opened Feb 27, 2026 by lichengzhe Loading…
Fix hybrid cache checkpoints for short conversations
#929 opened Feb 25, 2026 by omui-ai • Draft
Hybrid cache for Qwen3.5
#923 opened Feb 23, 2026 by dexloom Loading…
3 tasks
Add --q-override for per-layer quantization
#922 opened Feb 23, 2026 by spicyneuron Loading…
4
fix(types): improve type hints for mlx_lm.utils.load method
#919 opened Feb 22, 2026 by rahuliyer95 Loading…
5
Fix JSONDecodeError in qwen3_coder tool parser
#913 opened Feb 20, 2026 by BrunoCerberus Loading…
1
Add MTP support for Step 3.5 Flash
#901 opened Feb 16, 2026 by janhilgard Loading…
5 of 6 tasks
fix: handle tool call parse errors gracefully in server
#899 opened Feb 16, 2026 by shanemmattner Loading…
2
Add per-sequence trim support to BatchKVCache
#873 opened Feb 11, 2026 by 0xDaizz Loading…
Fix dynamic_quant for MoE and VL models
#870 opened Feb 10, 2026 by Taderich73 Loading…
3 tasks done
Add support for ministral3 and mistral3 model types
#860 opened Feb 7, 2026 by sealad886 Loading…
2
server: log bad tool_calls from model
#850 opened Feb 6, 2026 by percontation Loading…
refactor: use time.perf_counter() for duration measurements
#848 opened Feb 6, 2026 by m92y Loading…
feat: enhance chat CLI with readline history, line editing, and distributed support
#841 opened Feb 3, 2026 by Vlor999 Loading…
1
12
feat: Add LLaDA 1.0/2.0 Diffusion Model Support
#804 opened Jan 24, 2026 by akkikiki • Draft
3
Add support for LongCat ZigZag Attention
#802 opened Jan 23, 2026 by kernelpool Loading…
2
moved more activation functions to activations module
#774 opened Jan 19, 2026 by Goekdeniz-Guelmez Loading…
7
sov30_moe sparse moe blocks
#734 opened Jan 6, 2026 by rachittibrewal-sarvam • Draft
1
Add Zamba2
#724 opened Jan 4, 2026 by proazr Loading…
1
ProTip! What’s not been updated in a month: updated:<2026-02-07.