Skip to content

Pull requests: NVIDIA/TransformerEngine

New pull request New
85 Open 1,558 Closed
85 Open 1,558 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

NVFP4 with support for fusible ops wontfix This will not be worked on
#2190 opened Sep 19, 2025 by timmoon10 • Draft
1
[Feature] Enable rope application with offsets for training
#2188 opened Sep 19, 2025 by sudhakarsingh27 Loading…
1 of 13 tasks
2
[Feat] Draft: support offloading activation
#2187 opened Sep 18, 2025 by lhb8125 Loading…
13 tasks
[JAX] Local-Amax for Current-Scaling 2.8.0
#2183 opened Sep 17, 2025 by mingxu1067 Loading…
1 of 13 tasks
16
[Core][PyTorch] NVFP4 recipe 2.8.0
#2177 opened Sep 16, 2025 by ksivaman Loading…
3 of 13 tasks
@ksivaman
22
Context Parallel integration tests with a transformer layer: BSHD and THD + CP
#2176 opened Sep 16, 2025 by jomitchellnv Loading…
7 of 13 tasks
[PyTorch Debug] Add nvdlfw-inspect to dependencies
#2173 opened Sep 15, 2025 by pggPL • Draft
7 tasks done
1
[QA] Add pytest xml report for all tests in qa folder that use pytest
#2169 opened Sep 10, 2025 by shengfangd Loading…
7 of 13 tasks
7
blockwise fp8 weight memory optimization: on-demand columnwise fp8 weight creation
#2168 opened Sep 10, 2025 by skydoorkai Loading…
7 of 13 tasks
4
[JAX] Restore Shardy Rule with CompoundFactor
#2167 opened Sep 9, 2025 by phu0ngng • Draft
1 of 13 tasks
[JAX] CollectiveGemm 2.8.0
#2166 opened Sep 9, 2025 by phu0ngng Loading…
8 of 13 tasks
2
Fix issue with RNG state shape
#2164 opened Sep 8, 2025 by epwalsh Loading…
5 of 13 tasks
1
3
Support for Swiglu Activation used in GPT OSS
#2161 opened Sep 8, 2025 by vthumbe1503 Loading…
8 of 13 tasks
17
Add support for the FP8 Block Scaling (ie. Deepseek) recipe on Blackwell 2.8.0
#2157 opened Sep 5, 2025 by janekb04 Loading…
5 of 13 tasks
6
[Common][PyTorch][Rework] PDL for Quantization
#2150 opened Sep 4, 2025 by yaox12 Loading…
1 of 13 tasks
9
[PyTorch] Add sink attention support from cuDNN 2.8.0
#2148 opened Sep 2, 2025 by cyanguwa Loading…
8 of 13 tasks
1
7
[PyTorch] CPU Overhead Micro-optimizations
#2146 opened Sep 2, 2025 by zhongbozhu Loading…
13 tasks
@zhongbozhu
10
[main][feature][under updating]adapt for offload activation
#2145 opened Sep 2, 2025 by GeYuhong Loading…
1 of 13 tasks
[PyTorch] Add record_stream and untyped_storage func op in QuantizedTensor
#2144 opened Sep 2, 2025 by xiaoxi-wangfj Loading…
1 of 13 tasks
[PyTorch Debug] Support precision debug tools for fp8 model parameters.
#2141 opened Sep 1, 2025 by pggPL Loading…
8 of 13 tasks
1
3
ci: Build and attach bdist wheels to release page
#2138 opened Aug 29, 2025 by ko3n1g Loading…
13 tasks
[PyTorch Debug] Add max_blockwise_dynamic_range stats
#2137 opened Aug 29, 2025 by pggPL Loading…
8 of 13 tasks
1
Adds dst.dtype information in copy_ method of quantized tensors.
#2120 opened Aug 26, 2025 by zobeideThePlayer Loading…
3 of 13 tasks
5
[PyTorch Debug] Draft of tutorial for custom precision testing.
#2112 opened Aug 25, 2025 by pggPL • Draft
1 of 13 tasks
[PyTorch Debug] Fix issue with microbatching + debug value caching
#2108 opened Aug 25, 2025 by pggPL Loading…
8 of 13 tasks
6
ProTip! What’s not been updated in a month: updated:<2025-08-19.