]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: ecbcb7e)
tests : model metadata loading from huggingface (#19796)
authorBartowski <redacted>
Sat, 28 Feb 2026 09:44:38 +0000 (04:44 -0500)
committerGitHub <redacted>
Sat, 28 Feb 2026 09:44:38 +0000 (10:44 +0100)
commitd979f2b176217a062c57c24355db8c1dc3dfc7d5
tree1596725299564d15f94b80546994921f8f6c3f0etree
parentecbcb7ea9d3303097519723b264a8b5f1e977028commit | diff
tests : model metadata loading from huggingface (#19796)

* Add model metadata loading from huggingface for use with other tests

* Add incremental chunking instead of full redownload, fix caching issue and add warning when it fails

* Add support for split models, load metadata from each individual split file, also avoid mmproj

* Code cleanup, revert incremental downloading

* Only compile when cpp-httplib has SSL support

* Fix formatting
tests/CMakeLists.txt diff | blob | history
tests/gguf-model-data.cpp [new file with mode: 0644] blob
tests/gguf-model-data.h [new file with mode: 0644] blob
tests/test-gguf-model-data.cpp [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp