Skip to content

Pull requests: quic/efficient-transformers

New pull request New
41 Open 856 Closed
41 Open 856 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

revert(export): Revert proxy-only ONNX transform gating and restore default export behavior 1.21.0
#912 opened Apr 10, 2026 by vbaddi Contributor Loading…
@abhishek-singh591
MLA perf
#910 opened Apr 8, 2026 by quic-mamta Contributor Loading…
Adding multi_specializations_frames
#909 opened Apr 7, 2026 by mohiso22 Contributor Loading…
@mohiso22
14
Enable causal blocking
#908 opened Apr 6, 2026 by kdulla Contributor Loading…
2 tasks done
2
[QEff. Finetuning] Adding TP+DDP support in HF Trainer stack
#907 opened Apr 6, 2026 by smedhe Contributor • Draft
feat: Enable benchmark-mode module inventory/export across all CausalLM architectures enhancement New feature or request
#906 opened Apr 3, 2026 by vbaddi Contributor Loading…
2
feat(Llama GQA): optional shared-KV matmul rewrite for attention. WIP: (env-gated)
#905 opened Apr 3, 2026 by vbaddi Contributor • Draft
feat: Named graph specializations in specializations.json (Prefill/Decode/Vision/Encoder/Embedding) enhancement New feature or request
#904 opened Apr 3, 2026 by vbaddi Contributor Loading…
@vbaddi
[CI: NIGHTLY] Three-way Execution full layer, few layer, and dummy
#903 opened Apr 2, 2026 by abukhoy Contributor Loading…
qwen3_5_linear_attn
#901 opened Apr 1, 2026 by mohiso22 Contributor Loading…
[Experimental] Prefill opt moe
#895 opened Mar 30, 2026 by ochougul Contributor • Draft
Dependency Package update
#890 opened Mar 26, 2026 by quic-hemagnih Contributor Loading…
Add old stack logging support to new stack
#889 opened Mar 26, 2026 by quic-abhamidi Loading…
16
Added performance analysis as a feature with AutoModelForCausalLM
#888 opened Mar 25, 2026 by ochougul Contributor • Draft
@ochougul
Merge ft_experimental_v1 branch to main fine-tuning ready for review
#887 opened Mar 25, 2026 by quic-akuruvil Contributor Loading…
@quic-akuruvil @quic-swatia
@tchawada @smedhe
Unit test workflow
#878 opened Mar 23, 2026 by quic-rishinr Contributor Loading…
4
rebase(transformers): align modeling wrappers, cache_utils and other changes to v5.3.0 and restore PyTorch/ORT parity enhancement New feature or request
#876 opened Mar 21, 2026 by vbaddi Contributor • Draft
2 tasks
@vbaddi @asmigosw
1
Undo deepstack_features based changes for Qwen3VL and Qwen3VL_MOE models
#869 opened Mar 18, 2026 by quic-dhirajku Contributor • Draft
MLA : update attention in fused_forward, head blocking and add prefillonly transform
#857 opened Mar 16, 2026 by quic-mamta Contributor Loading…
Added NPI files
#837 opened Mar 9, 2026 by asmigosw Contributor Loading…
1
[Nightly CI]: Creating CI Pipeline for Nightly Build
#828 opened Mar 5, 2026 by abukhoy Contributor • Draft
Added fp16/bf16 based export and compile support for VLMs
#819 opened Mar 2, 2026 by asmigosw Contributor Loading…
48
FirstCache for Diffusers
#803 opened Feb 23, 2026 by quic-amitraj Contributor • Draft
FT-CI enabled on torch-qaic-env fine-tuning
#799 opened Feb 18, 2026 by quic-akuruvil Contributor Loading…
@quic-akuruvil
2
Add support for num_crops and valid_size from vLLM
#796 opened Feb 17, 2026 by quic-vargupt Contributor Loading…
ProTip! Type g i on any issue or pull request to go back to the issue listing page.