Skip to content

Pull requests: quic/efficient-transformers

New pull request New
88 Open 1,214 Closed
88 Open 1,214 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

Updated numpy and protobug version
#1321 opened Sep 10, 2026 by quic-rishinr Contributor Loading…
[Qwen/Qwen3-VL-235B-A22B] Change vision batch size to 1
#1320 opened Sep 9, 2026 by quic-sanising Contributor Loading…
Ci/llm first impact enable in CI
#1319 opened Sep 9, 2026 by ochougul Contributor Loading…
1
Minimax m3 mask rope cache
#1318 opened Sep 9, 2026 by ochougul Contributor • Draft
Minimax m3 weight free export
#1317 opened Sep 9, 2026 by amarquic Contributor Loading…
nit(qwen35): Optimize Qwen3.5-MoE folded decode execution 1.23 Release 1.23 Features enhancement New feature or request
#1315 opened Sep 9, 2026 by vbaddi Contributor Loading…
@vbaddi @mohiso22
1
Enable Dynamo KV-blocking Loop export for kv_headpar
#1314 opened Sep 8, 2026 by quic-abhamidi Loading…
fix: return decode QPC key for CB decode-only compile
#1313 opened Sep 8, 2026 by ochougul Contributor Loading…
Wire QEffGptOssMLP into shared MoEBlockMixin
#1312 opened Sep 8, 2026 by Shrubabati7 Loading…
6
Enable wight-free export for VLM
#1311 opened Sep 8, 2026 by quic-amitraj Contributor • Draft
1
Enable weight-free MXFP6 DQ export
#1310 opened Sep 7, 2026 by ochougul Contributor Loading…
fix(write_io): capture complete generate() I/O traces for LLM and VLM models
#1309 opened Sep 7, 2026 by talnufai Contributor Loading…
1
Pr 94 99 01
#1307 opened Sep 3, 2026 by butanehi Loading…
[feature] Enable Dynamic Batching
#1305 opened Sep 2, 2026 by eplatero97 Contributor Loading…
1
COMBINING: Adding latest forward substitution in Gated Deltanet AND: feat(0831): Localize constant ReduceSum axes inside ONNX subfunctions
#1304 opened Sep 2, 2026 by butanehi Loading…
Move pytorch transforms from init method to transform method
#1303 opened Sep 2, 2026 by quic-mamta Contributor • Draft
@quic-mamta
Fix GPT-OSS blocked prefill cache reads and head-par prefill blocking
#1302 opened Sep 2, 2026 by smedhe Contributor • Draft
2
Generic prefill blocking support for all models qeff.blocking
#1301 opened Sep 1, 2026 by kdulla Contributor Loading…
@kdulla
1
Add support for GLM-4.7-Flash (glm4_moe_lite)
#1300 opened Sep 1, 2026 by Knightkolla Loading…
1
Adding latest forward substitution in Gated Deltanet
#1299 opened Sep 1, 2026 by mohiso22 Contributor Loading…
18
Unify multimodal example scripts
#1298 opened Sep 1, 2026 by quic-mamta Contributor • Draft
@quic-mamta
fix(gemma3): Fix OCR regression caused by RoPE ONNX aliasing
#1295 opened Aug 31, 2026 by prabeesh-1 Loading…
2
feat(0831): Localize constant ReduceSum axes inside ONNX subfunctions 1.23 Release 1.23 Features enhancement New feature or request
#1294 opened Aug 31, 2026 by vbaddi Contributor Loading…
@vbaddi
3
Fix Gemma4 RoPE handling
#1293 opened Aug 31, 2026 by nsivalin Contributor • Draft
1
fix(gemma4): Fix OCR regression caused by RoPE ONNX aliasing
#1292 opened Aug 30, 2026 by prabeesh-1 Loading…
2
ProTip! What’s not been updated in a month: updated:<2026-08-10.