Skip to content

Pull requests: lablup/mlxcel

New pull request New
12 Open 926 Closed
12 Open 926 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

feat(server): send sampled video frames as ordered images area:cli Command-line interface / CLI flags area:inference Generation, sampling, decoding (incl. speculative, DRY) modeltype:vlm Vision-language model priority:medium Medium priority status:done Completed type:enhancement New features, capabilities, or significant additions
#1759 opened Sep 10, 2026 by inureyes Member Loading…
1
feat(models): add the IQuest-Coder Loop two-pass decoder (iquestloopcoder) arch:dense Dense transformer decoder area:docs User and developer documentation area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata modelsize:medium 4-bit checkpoint 10-50GB; comfortable on a 128GB dev box modeltype:text Text-only language model priority:medium Medium priority status:done Completed type:enhancement New features, capabilities, or significant additions
#1757 opened Sep 10, 2026 by inureyes Member Loading…
7 of 8 tasks
1
feat(speculative): GLM-4.7-Flash (glm4_moe_lite) MTP drafter arch:moe Sparse mixture-of-experts decoder area:cli Command-line interface / CLI flags area:core mlxcel-core: MLX FFI, primitives, KV cache, layers area:docs User and developer documentation area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata modelsize:medium 4-bit checkpoint 10-50GB; comfortable on a 128GB dev box modeltype:text Text-only language model priority:low Low priority status:done Completed type:enhancement New features, capabilities, or significant additions
#1753 opened Sep 10, 2026 by inureyes Member Loading…
1
feat: surface reasoning-only truncation on the chat completions API
#1746 opened Sep 10, 2026 by bebekim Loading…
11 of 12 tasks
1
1
chore(build): include benches/ in insert_apache_header.py target roots
#1728 opened Sep 9, 2026 by saitejagirada Loading…
2 tasks done
1
1
feat(vlm): add vision feature caching for LFM2-VL area:models Model architectures, weights, loading, metadata modeltype:vlm Vision-language model priority:medium Medium priority status:review Under review type:enhancement New features, capabilities, or significant additions
#1135 opened Aug 13, 2026 by MrFelt0385 Loading…
2
feat(xla): add Gemma3n audio execution area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:low Low priority status:blocked Blocked by dependencies or other issues type:enhancement New features, capabilities, or significant additions
#922 opened Jul 24, 2026 by inureyes Member • Draft
1
4
feat(xla): add Qwen3-VL DeepStack vision path area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:low Low priority status:review Under review type:enhancement New features, capabilities, or significant additions
#921 opened Jul 24, 2026 by inureyes Member • Draft
35 of 43 tasks
1
4
feat: add Qwen2.5-VL OpenXLA vision path area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:low Low priority status:review Under review type:enhancement New features, capabilities, or significant additions
#920 opened Jul 24, 2026 by inureyes Member • Draft
1
4
feat: add Youtu-VL XLA vision and MLA inference (#872) area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:low Low priority status:review Under review type:enhancement New features, capabilities, or significant additions
#919 opened Jul 24, 2026 by inureyes Member • Draft
1
2
feat(xla): support Molmo sparse additive visual embeddings area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:low Low priority status:review Under review type:enhancement New features, capabilities, or significant additions
#918 opened Jul 24, 2026 by inureyes Member • Draft
1
8
feat: add Gemma3 VLM OpenXLA path (#869) area:inference Generation, sampling, decoding (incl. speculative, DRY) area:models Model architectures, weights, loading, metadata priority:medium Medium priority status:review Under review type:enhancement New features, capabilities, or significant additions
#917 opened Jul 24, 2026 by inureyes Member • Draft
23 of 27 tasks
1
3
ProTip! What’s not been updated in a month: updated:<2026-08-10.