-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Pull requests: ml-explore/mlx-lm
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix incorrect assumption when parsing tool_calls[].function.arguments
#1904
opened Sep 20, 2026 by
LxYuan0420
Contributor
Loading…
Add DeepSeek-V4.1 model
new_model
#1895
opened Sep 16, 2026 by
Goekdeniz-Guelmez
Contributor
Loading…
server: --allowed-models to restrict which model paths a request may load
#1893
opened Sep 15, 2026 by
mrcushen-arch
Loading…
Swallow missing-stream errors when BatchGenerator is torn down off its GPU thread
#1890
opened Sep 15, 2026 by
gyanu2507
Contributor
Loading…
Tolerate a null rope_scaling config when constructing PhiMoE models
#1889
opened Sep 15, 2026 by
gyanu2507
Contributor
Loading…
Chunkwise-parallel gated delta rule for training
#1870
opened Sep 10, 2026 by
adityak74
Contributor
Loading…
Report the matched stop sequence on GenerationBatch.Response
#1869
opened Sep 9, 2026 by
mloiterman
Contributor
Loading…
Parse gpt-oss's harmony channel format in non-streaming completions
#1867
opened Sep 9, 2026 by
eejd
Loading…
Support uneven tensor-parallel sharding in shard() (GQA/MLA-aware)
#1863
opened Sep 8, 2026 by
twallgren
Contributor
Loading…
Add global scale for nvfp4 quantized MoEs
#1844
opened Sep 4, 2026 by
nastya236
Collaborator
Loading…
fix: Speed up multi-token decode by computing the absorbed MLA threshold
#1817
opened Sep 2, 2026 by
stooit
Loading…
Add DeepSeek-V4-Flash without MTP (yet)
new_model
#1797
opened Aug 28, 2026 by
nh13
Contributor
Loading…
Add Qwen3.8-Flash-Next (qwen4_exp) model support
await verification
This pull request is non-trivial and requires a human expert to verify its correctness.
#1788
opened Aug 26, 2026 by
eauchs
Loading…
Add Intern-S2-Mobius (interns2_mobius) model
new_model
#1771
opened Aug 21, 2026 by
nightscape
•
Draft
Add M-RoPE support so vision embeddings keep their grid positions
enhancement
#1768
opened Aug 21, 2026 by
lpalbou
Contributor
Loading…
Fix deepseek_v32 Indexer evicting attention sinks from sparse top-k
bug
#1552
opened Jul 11, 2026 by
robertlangdonn
Contributor
Loading…
5 tasks done
Add Nemotron-H Puzzle model support
await response
This pull request is waiting for response from the author.
new_model
#1535
opened Jul 10, 2026 by
sxuff
Loading…
Keep the server generation loop alive when a batched request fails
bug
#1513
opened Jul 10, 2026 by
rajanshxrma
Contributor
Loading…
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.