ggml-cuda: key the CUDA graph cache by shape

5a08a71

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

Pull Request Labeler on: pull_request_target

labeler

flake8 Lint on: pull_request 1

Lint

CI (self-hosted) on: pull_request 14

gpu-cuda

gpu-rocm

gpu-vulkan-nvidia-cm

gpu-vulkan-nvidia-cm2

gpu-webgpu-nvidia

gpu-metal

gpu-webgpu-apple

gpu-vulkan-apple

gpu-vulkan-intel-linux

gpu-vulkan-intel-windows

gpu-openvino-low-perf

cpu-x64-high-perf

cpu-arm64-high-perf-graviton4

cpu-arm64-graviton4-kleidiai

Server on: pull_request

ubuntu

windows

CI (wasm) on: pull_request

ubuntu-webgpu

HIP quality check on: pull_request

ubuntu-22-hip-quality-check

CI (CUDA, ubuntu) on: pull_request

cuda

hip

musa

Python check requirements.txt on: pull_request 1

check-requirements

Check Pre-Tokenizer Hashes on: pull_request 1

pre-tokenizer-hashes

Python Type-Check on: pull_request 1

python type-check

CI (cpu) on: pull_request 6

build-cmake-pkg / linux

ubuntu (x64, ubuntu-22.04)

ubuntu (arm64, ubuntu-24.04-arm)

windows / x64-cpu-static

windows / x64-openblas

windows / arm64

labeler

succeeded Sep 2, 2026 in 19s

Loading

Re-running jobs...