Skip to content

Navigation Menu

Appearance settings

View all features
View all solutions
View all resources
- COMMUNITY
  GitHub SponsorsFund open source developers
- PROGRAMS
  Security Lab
  Maintainer Community
  Accelerator
  GitHub Stars
  Archive Program
- REPOSITORIES
  Topics
  Trending
  Collections
- ENTERPRISE SOLUTIONS
  Enterprise platformAI-powered developer platform
- AVAILABLE ADD-ONS
  GitHub Advanced SecurityEnterprise-grade security features
  Copilot for BusinessEnterprise-grade AI features
  Premium SupportEnterprise-grade 24/7 support
Pricing

Search code, repositories, users, issues, pull requests...

Search

Clear

Search syntax tips

Provide feedback

We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Saved searches

Use saved searches to filter your results more quickly

Name

Query

To see all available qualifiers, see our documentation.

Appearance settings

You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session.

Dismiss alert

Uh oh!

There was an error while loading. Please reload this page.

pytorch / TensorRT Public

Notifications You must be signed in to change notification settings
Fork 411
Star 3k

Code
Issues 266
Pull requests 77
Discussions
Actions
Projects
Wiki
Security and quality
Insights

Additional navigation options

Code
Issues
Pull requests
Discussions
Actions
Projects
Wiki
Security and quality
Insights

fix(executorch): support KV-cache aliased I/O in the TensorRT delegate - #4445

#4445

Open

Conarnar wants to merge 2 commits into

pytorch:mainpytorch/TensorRT:mainfrom

Conarnar:fix/executorch-kv-alias-bindingsConarnar/TensorRT:fix/executorch-kv-alias-bindingsCopy head branch name to clipboard

Conversation Commits2 (2)Checks Files changed

Open

fix(executorch): support KV-cache aliased I/O in the TensorRT delegate#4445
Conarnar wants to merge 2 commits into
pytorch:mainfrom
Conarnar:fix/executorch-kv-alias-bindings

feat(executorch): caller-owned KV-cache for the TensorRT delegate

2312f42

Select commit

Loading

Failed to load commit list.

Uh oh!

There was an error while loading. Please reload this page.

Sign in for the full log view

Labeler on: pull_request_target 1

label

Meta CLA

Meta CLA Check

Lint Code on: pull_request 2

C++ Linting

Python Linting

RTX - Build Linux aarch64 wheels on: pull_request 3

generate-matrix / generate

filter-matrix

RTX - Build SBSA torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-aarch64-aarch64-true-false

RTX - Build SBSA torch-tensorrt whl package / upload-wheel----false / upload-wheel-py3_10-cuda-aarch6413_0-aarch64

RTX - Build and test Windows wheels on: pull_request 23

generate-matrix / generate

filter-matrix

substitute-runner

RTX - Build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_0

RTX - Build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_2

RTX - Build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_0

RTX - Build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_2

L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu130

L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu132

L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu130

L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu132

L0 core python tests / L0-core-python-tests--3.10-cu130

L0 core python tests / L0-core-python-tests--3.10-cu132

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

Build and test Windows wheels on: pull_request 29

generate-matrix / generate

filter-matrix

substitute-runner

Build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_0

Build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_2

Build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_0

Build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_2

L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu130

L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu132

L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu130

L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu132

L0 core python tests / L0-core-python-tests--3.10-cu130

L0 core python tests / L0-core-python-tests--3.10-cu132

L0 torchscript tests / L0-torchscript-tests--3.10-cu130

L0 torchscript tests / L0-torchscript-tests--3.10-cu132

matrix.display-name

matrix.display-name

matrix.display-name

L1 torchscript tests

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

matrix.display-name

Python-only build and test Windows wheels on: pull_request 11

generate-matrix / generate

filter-matrix

substitute-runner

Python-only build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_0

Python-only build Windows torch-tensorrt whl package / build-wheel-py3_10-cuda13_2

Python-only build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_0

Python-only build Windows torch-tensorrt whl package / upload / upload-wheel-py3_10-cuda13_2

Python-only dynamo runtime tests / python-only-dynamo-runtime-tests--3.10-cu130

Python-only dynamo runtime tests / python-only-dynamo-runtime-tests--3.10-cu132

Python-only build and test Linux x86_64 wheels on: pull_request 33

generate-matrix / generate

filter-matrix

Python-only build Linux x86_64 torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-x86_64-false-false

Python-only build Linux x86_64 torch-tensorrt whl package / build-wheel-3.10-cu132-cuda-x86_64-false-false

Python-only build Linux x86_64 torch-tensorrt whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_0

Python-only build Linux x86_64 torch-tensorrt whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_2

Python-only dynamo runtime tests / python-only-dynamo-runtime-tests--3.10-cu130

Python-only dynamo runtime tests / python-only-dynamo-runtime-tests--3.10-cu132

RTX - Python-only build and test Linux x86_64 wheels on: pull_request 44

generate-matrix / generate

filter-matrix

RTX - Python-only build Linux x86_64 torch-tensorrt-rtx whl package / build-wheel-3.10-cu130-cuda-x86_64-true-false

RTX - Python-only build Linux x86_64 torch-tensorrt-rtx whl package / build-wheel-3.10-cu132-cuda-x86_64-true-false

RTX - Python-only build Linux x86_64 torch-tensorrt-rtx whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_0

RTX - Python-only build Linux x86_64 torch-tensorrt-rtx whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_2

RTX - Python-only dynamo runtime tests / rtx-python-only-dynamo-runtime-tests--3.10-cu130

RTX - Python-only dynamo runtime tests / rtx-python-only-dynamo-runtime-tests--3.10-cu132

Build and test Linux aarch64 wheels on: pull_request 3

generate-matrix / generate

filter-matrix

Build SBSA torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-aarch64-aarch64-false-false

Build SBSA torch-tensorrt whl package / upload-wheel----false / upload-wheel-py3_10-cuda-aarch6413_0-aarch64

RTX - Python-only build and test Windows wheels on: pull_request 11

generate-matrix / generate

filter-matrix

substitute-runner

RTX - Python-only build Windows torch-tensorrt-rtx whl package / build-wheel-py3_10-cuda13_0

RTX - Python-only build Windows torch-tensorrt-rtx whl package / build-wheel-py3_10-cuda13_2

RTX - Python-only build Windows torch-tensorrt-rtx whl package / upload / upload-wheel-py3_10-cuda13_0

RTX - Python-only build Windows torch-tensorrt-rtx whl package / upload / upload-wheel-py3_10-cuda13_2

RTX - Python-only dynamo runtime tests / rtx-python-only-dynamo-runtime-tests--3.10-cu130

RTX - Python-only dynamo runtime tests / rtx-python-only-dynamo-runtime-tests--3.10-cu132

Python-only build Linux aarch64 wheels on: pull_request 3

generate-matrix / generate

filter-matrix

Python-only build Linux aarch64 torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-aarch64-aarch64-false-false

Python-only build Linux aarch64 torch-tensorrt whl package / upload-wheel----false / upload-wheel-py3_10-cuda-aarch6413_0-aarch64

Build and test Linux x86_64 wheels on: pull_request 65

core / generate-matrix / generate

core / filter-matrix

core / Build Linux x86_64 torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-x86_64-false-false

core / Build Linux x86_64 torch-tensorrt whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_0

core / executorch-static-build / select-matrix

core / executorch-static-build / build / executorch-static-build--3.10-cu130

core / L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu130

core / L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu130

core / L0 core python tests / L0-py-core-tests--3.10-cu130

core / L0 torchscript tests / L0-torchscript-tests--3.10-cu130

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / collect-results

CI / Linux x86_64

RTX - Build and test Linux x86_64 wheels on: pull_request 43

core / generate-matrix / generate

core / filter-matrix

core / RTX - Build Linux x86_64 torch-tensorrt whl package / build-wheel-3.10-cu130-cuda-x86_64-true-false

core / RTX - Build Linux x86_64 torch-tensorrt whl package / upload-wheel----false / upload-manywheel-py3_10-cuda13_0

core / executorch-static-build

core / L0 dynamo converter tests / L0-dynamo-converter-tests--3.10-cu130

core / L0 dynamo core tests / L0-dynamo-core-tests--3.10-cu130

core / L0 core python tests / L0-py-core-tests--3.10-cu130

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / matrix.display-name

core / collect-results

CI / Linux x86_64 (RTX)

Annotations

1 warning

Complete job

Node.js 20 is deprecated. The following actions target Node.js 20 but are being forced to run on Node.js 24: actions/checkout@v4. For more information see: https://github.blog/changelog/2025-09-19-deprecation-of-node-20-on-github-actions-runners/

label

succeeded Aug 1, 2026 in 14s

Loading

Re-running jobs...

Footer

© 2026 GitHub, Inc.

Footer navigation

Terms
Privacy
Security
Status
Community
Docs
Contact

You can’t perform that action at this time.