]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 5ec717d)
ggml-webgpu: Enable NVIDIA self-hosted CI (#22976)
authorReese Levine <redacted>
Thu, 14 May 2026 16:41:32 +0000 (09:41 -0700)
committerGitHub <redacted>
Thu, 14 May 2026 16:41:32 +0000 (09:41 -0700)
commit834a243664114487f99520370a7a7b00fc7a486f
tree278ba1ea69c0c34c95aea5368895c435b4638183tree
parent5ec717d1256e34558a44dc09adf1e6e16f2e2682commit | diff
ggml-webgpu: Enable NVIDIA self-hosted CI (#22976)

* Enabel nvidia ci for webgpu

* Address precision issues

* fix placement

* Relax more set_rows and div

* Try relaxing all f16

* formatting and naming

* Add comment explaining max_nmse_err logic

Added comment referencing pull request for clarification.
.github/workflows/build-self-hosted.yml diff | blob | history
tests/test-backend-ops.cpp diff | blob | history
Packaging of ggml-org/llama.cpp