-
Notifications
You must be signed in to change notification settings - Fork 364
Insights: pytorch/TensorRT
Overview
Could not load contribution data
Please try again later
9 Pull requests merged by 2 people
-
Add Flux fp4 support
#3689 merged
Jul 18, 2025 -
cherry-pick 3686: upgrade tensorrt from 10.11 to 10.12
#3690 merged
Jul 17, 2025 -
upgrade tensorrt from 10.11 to 10.12
#3686 merged
Jul 16, 2025 -
fix windows build failure: add /utf-8
#3684 merged
Jul 16, 2025 -
cherry pick 3680: fix refit test bug
#3687 merged
Jul 16, 2025 -
cherry pick 3685: disable jetson build in ci
#3688 merged
Jul 16, 2025 -
Fixed the CI problem
#3680 merged
Jul 15, 2025 -
disable jetpack build for now
#3685 merged
Jul 15, 2025 -
remove tensorrt as build dependency
#3681 merged
Jul 14, 2025
2 Pull requests opened by 2 people
-
Index converter dynamic cases fix
#3694 opened
Jul 17, 2025 -
cherry pick 3689 to 2.8 release:flux fp4
#3696 opened
Jul 18, 2025
2 Issues closed by 1 person
-
🐛 [Bug] torchtrt.dynamo.compile produces nan values
#3596 closed
Jul 16, 2025 -
🐛 [Bug] dynamo.compile fails in 2.9.0 nightly when mqt quantization is used
#3659 closed
Jul 16, 2025
4 Issues opened by 3 people
-
🐛 [Bug] Wan2.1 SDPA in Torch-TRT is slower than SDPA in ONNX as batch size (num_frames) grow
#3695 opened
Jul 17, 2025 -
❓ [Question] How to understand the value of this project
#3691 opened
Jul 17, 2025 -
❓ [Question] HELP:dynamic shape of offset and input is not supported in aten_ops_embedding_bag converter
#3683 opened
Jul 15, 2025 -
🐛 [Bug] TorchTRT is 2x slower than ONNX on SD3.5. Same Kernel has different performance
#3682 opened
Jul 11, 2025
12 Unresolved conversations
Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.
-
Upgrade perf_run script to support TRT 10 and fix some issues
#3650 commented on
Jul 18, 2025 • 2 new comments -
🐛 [Bug] group_norm implementation may slows down SD vae
#3648 commented on
Jul 11, 2025 • 0 new comments -
🐛 [Bug] EfficientNet Conv is not fused with Activations (SiLU)
#3678 commented on
Jul 11, 2025 • 0 new comments -
🐛 [Bug] compiled F.grid_sample produces wrong result
#3674 commented on
Jul 14, 2025 • 0 new comments -
🐛 [Bug] Large Accuracy Issue
#3626 commented on
Jul 15, 2025 • 0 new comments -
TRT-LLM loading mechanism tool
#3398 commented on
Jul 18, 2025 • 0 new comments -
Cherrypick of PR 3570
#3662 commented on
Jul 18, 2025 • 0 new comments -
Cherrypick of PR 3513
#3664 commented on
Jul 18, 2025 • 0 new comments -
chore: cherry pick of bf16 cast PR (3643)
#3666 commented on
Jul 18, 2025 • 0 new comments -
chore: cherry pick of KV cache PR (3527)
#3667 commented on
Jul 18, 2025 • 0 new comments -
Disabled silu decomposition cast
#3677 commented on
Jul 17, 2025 • 0 new comments -
tensorrt_rtx try on
#3679 commented on
Jul 18, 2025 • 0 new comments