]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 74a6d92)
Allow "quantizing" to f16 and f32 (#1787)
authorKerfuffle <redacted>
Tue, 13 Jun 2023 10:23:23 +0000 (04:23 -0600)
committerGitHub <redacted>
Tue, 13 Jun 2023 10:23:23 +0000 (04:23 -0600)
commit74d4cfa3438cb58bd177eed30014e6588694aaa8
tree8252d459812376411bcf8508ad7d1f059670601atree
parent74a6d922f12ccfe16b0c265f43be8978c6f25e98commit | diff
Allow "quantizing" to f16 and f32 (#1787)

* Allow "quantizing" to f16 and f32

Fix an issue where quantizing didn't respect LLAMA_NO_K_QUANTS

Add brief help to the list of quantization types in the quantize tool

Ignore case for quantization type arguments in the quantize tool
Makefile diff | blob | history
examples/quantize/quantize.cpp diff | blob | history
ggml.c diff | blob | history
llama.cpp diff | blob | history
Packaging of ggml-org/llama.cpp