Skip to content

Issues: NVIDIA/TransformerEngine

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

176 Open 225 Closed
176 Open 225 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
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

when to support flashattention2.7.3+
#1503 opened Feb 24, 2025 by leo-ztjht
Question about the performace of GroupedLinear
#1499 opened Feb 20, 2025 by XLzed
1
Float8Quantizer::create_tensor calculates scale_inv instead of creating an empty buffer performance
#1491 opened Feb 18, 2025 by yaox12
3
MXFP8 backward
#1490 opened Feb 18, 2025 by hanhanpp
@ksivaman
4
qwen1.5-0.5B failed to save model with huggingface transformers bug Something isn't working
#1482 opened Feb 13, 2025 by xinpengzz
2
When flash-attn >2.6.1, use context parallel will cause error
#1467 opened Feb 9, 2025 by south-ocean
@cyanguwa
1
How to set NVTE_FWD/BWD_LAYERNORM_SM_MARGIN?
#1459 opened Feb 5, 2025 by cailun01
1
If no Windows support is planned for 5090 as it wasn't for 4090, pass along to corporate to mention this in advertisement
#1436 opened Jan 29, 2025 by NeedsMoar
4
HF Accelerate FP8 use more gpu memory then FP16 in training LLM
#1429 opened Jan 28, 2025 by Liufeiran123
1
transformer_engine.pytorch.distributed.checkpoint function only works with TE modules, instead of all Callables
#1423 opened Jan 27, 2025 by MaciejBalaNV
FP8 execution requires 2D input matrices with height divisible by 8 and width divisible by 16
#1422 opened Jan 25, 2025 by Liufeiran123
1
Deadline or schedule new update supporting blackwell and fp4?
#1421 opened Jan 24, 2025 by johnnynunez
1
Problem when install transformers_engine with nvcc11.8 and nvcc12.0
#1420 opened Jan 23, 2025 by chwenjun225
3
Questions about accuracy alignment between BF16 and FP8 question Further information is requested
#1419 opened Jan 22, 2025 by zigzagcai
2
Performance Discrepancy in FP8 vs. BF16 Training with NanoGPT
#1416 opened Jan 21, 2025 by wzzll123
2
Plans for block-wise FP8 quantization during training?
#1411 opened Jan 15, 2025 by beccohov
3
Questions on DotProductAttention API Usage in Flash Attention thd Mode
#1409 opened Jan 14, 2025 by pipSu
@cyanguwa
1
Fused attention error while running Nvidia Cosmos
#1407 opened Jan 13, 2025 by deepbeepmeep
3
Import fails when working from a TE directory good first issue Good for newcomers
#1400 opened Jan 10, 2025 by ksivaman
@KshitijLakhani
Installation stuck at 97%
#1399 opened Jan 10, 2025 by lorenzbaraldi
1
why close ag overlap when is_grad_enabled is False
#1398 opened Jan 10, 2025 by sallyjunjun
1
support new flash_attn_interface
#1392 opened Jan 7, 2025 by rgtjf
@cyanguwa
2
FP8 GEMM Kernels
#1391 opened Jan 6, 2025 by xiaoxiao26
How about the grouplinear?
#1386 opened Dec 26, 2024 by south-ocean
2
_NoopCatFunc in transformer layer bug Something isn't working
#1384 opened Dec 22, 2024 by robot-transformer
@KshitijLakhani
ProTip! Mix and match filters to narrow down what you’re looking for.