Skip to content

Pull requests: pytorch/TensorRT

New pull request New
79 Open 2,471 Closed
79 Open 2,471 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

Added dynamo-torchao test suite under the full CI lane cla signed component: tests Issues re: Tests
#4601 opened Aug 26, 2026 by micwill755 Collaborator • 5/5 Loading…
5 of 8 tasks
perf(executorch): opt-in shared per-device activation-scratch pool cla signed component: api [C++] Issues re: C++ API component: tests Issues re: Tests
#4600 opened Aug 26, 2026 by Conarnar Contributor Loading…
7 tasks done
9
Skip rebuilding FX forward after post-lowering and before adjacency split. cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests
#4599 opened Aug 26, 2026 by micwill755 Collaborator • 8/8 • Draft
4 tasks done
Defer FX graph stringification until DEBUG logging is enabled. cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests component: torch_compile
#4598 opened Aug 26, 2026 by micwill755 Collaborator • 7/8 • Draft
1 of 2 tasks
Skip executing large aliased constant folds that 4531 will not install. cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests
#4597 opened Aug 26, 2026 by micwill755 Collaborator • 6/8 • Draft
1 of 2 tasks
Apply Torch-TRT decomps during dynamo.trace so compile skips a second AOT retrace. cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests
#4596 opened Aug 26, 2026 by micwill755 Collaborator • 5/8 Loading…
3 tasks done
Added TorchAO MXFP4 (MXDynamicActivationMXWeightConfig) + Linear and FLUX examples cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: converters Issues re: Specific op converters component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests documentation Improvements or additions to documentation
#4595 opened Aug 26, 2026 by micwill755 Collaborator • 4/5 Loading…
6 of 8 tasks
seperate executorch build test in ci ci: full cla signed component: api [Python] Issues re: Python API
#4594 opened Aug 26, 2026 by lanluo-nvidia Collaborator Loading…
6 of 7 tasks
v2.14.0
Added TorchAO NVFP4 weight-only quantization + Linear and FLUX examples cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: converters Issues re: Specific op converters component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests documentation Improvements or additions to documentation
#4593 opened Aug 26, 2026 by micwill755 Collaborator • 3/5 Loading…
5 of 7 tasks
Added TorchAO INT4 weight-only quantization + Linear, FLUX, and Qwen examples cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: converters Issues re: Specific op converters component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: tests Issues re: Tests documentation Improvements or additions to documentation
#4590 opened Aug 26, 2026 by micwill755 Collaborator • 2/5 Loading…
1 of 3 tasks
Added TorchAO FP8 weight-only and static quantization + user guides cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: converters Issues re: Specific op converters component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes component: tests Issues re: Tests documentation Improvements or additions to documentation
#4589 opened Aug 26, 2026 by micwill755 Collaborator • 1/5 Loading…
5 of 7 tasks
2
fix(tests): download the shared torchvision checkpoints before pytest forks ci: full cla signed component: tests Issues re: Tests
#4587 opened Aug 26, 2026 by shoumikhin Contributor Loading… v2.14.0
fix(runtime): back a python engine runtime cache with the python handle ci: full cla signed component: api [Python] Issues re: Python API component: tests Issues re: Tests
#4585 opened Aug 26, 2026 by shoumikhin Contributor Loading… v2.14.0
Fix brittle layer count cla signed component: tests Issues re: Tests
#4576 opened Aug 25, 2026 by jloftin-nv Contributor Loading…
1
[WIP]make executorch work in python-only torch-tensorrt build ci: full cla signed component: api [Python] Issues re: Python API component: build system Issues re: Build system component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: runtime WIP Work is in progress, pull request should not be merged yet
#4574 opened Aug 25, 2026 by lanluo-nvidia Collaborator • Draft
7 tasks
v2.15.0
ci(executorch): run a coalesced TensorRT + CUDA program in the reference runner gate ci: full cla signed component: tests Issues re: Tests
#4572 opened Aug 24, 2026 by shoumikhin Contributor Loading…
6 of 7 tasks
v2.14.0
feat: apply_runtime_settings — permanent runtime config for AOT-loaded engines cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: runtime component: tests Issues re: Tests documentation Improvements or additions to documentation
#4570 opened Aug 24, 2026 by tp5uiuc Collaborator • Draft
6 tasks done
Ship only the TensorRT delegate in the ExecuTorch runtime wheel ci: nightly Run the nightly lane (all tiers incl. llm / kernels / distributed) on every push cla signed component: api [C++] Issues re: C++ API component: api [Python] Issues re: Python API component: build system Issues re: Build system component: tests Issues re: Tests documentation Improvements or additions to documentation
#4567 opened Aug 23, 2026 by shoumikhin Contributor Loading…
6 of 7 tasks
v2.15.0
Track the ExecuTorch pin and move it to new nightlies automatically ci: nightly Run the nightly lane (all tiers incl. llm / kernels / distributed) on every push cla signed component: api [Python] Issues re: Python API component: build system Issues re: Build system component: tests Issues re: Tests documentation Improvements or additions to documentation
#4566 opened Aug 23, 2026 by shoumikhin Contributor Loading…
6 of 7 tasks
v2.15.0
1
Fix strided slice_scatter KV cache updates cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: converters Issues re: Specific op converters component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: tests Issues re: Tests
#4550 opened Aug 22, 2026 by apbose Collaborator Loading… v2.15.0
fix: add CUDA 13 support for TRT-LLM plugins cla signed component: api [Python] Issues re: Python API component: tests Issues re: Tests
#4549 opened Aug 22, 2026 by apbose Collaborator Loading… v2.14.0
fix: fall back to PyTorch for ops TensorRT-RTX cannot serve on Turing (SM 7.5) cla signed component: api [Python] Issues re: Python API component: conversion Issues re: Conversion stage component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: tests Issues re: Tests
#4546 opened Aug 21, 2026 by tp5uiuc Collaborator • Draft
4 of 7 tasks
Cache post-lowering ATen graphs to skip decomp on warm compiles. cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: tests Issues re: Tests
#4540 opened Aug 20, 2026 by micwill755 Collaborator • 4/8 • Draft
2 tasks done
v2.15.0
1
Batch cheap FX cleanups and defer post_lowering recompile cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes
#4532 opened Aug 19, 2026 by micwill755 Collaborator • 3/8 Loading…
3 tasks
1
Skip installing large aliased constant-folded weights cla signed component: api [Python] Issues re: Python API component: core Issues re: The core compiler component: dynamo Issues relating to the `torch.compile` or `torch._dynamo.export` paths component: lowering Issues re: The lowering / preprocessing passes
#4531 opened Aug 19, 2026 by micwill755 Collaborator • 2/8 Loading…
3 tasks
1
ProTip! Add no:assignee to see everything that’s not assigned.