diff options
author | Georgi Gerganov <ggerganov@gmail.com> | 2023-04-20 20:35:53 +0300 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-04-20 20:35:53 +0300 |
commit | e0305ead3a072db9c08b35c9600c49273b38a4b5 (patch) | |
tree | 6352a4bb010b4e2aa7136bdb925d7a7a82e83fd6 /llama.h | |
parent | 6a9661ea5ad72166b700ae5e87976e4452499dda (diff) |
ggml : add Q4_3 quantization (#1082)
Diffstat (limited to 'llama.h')
-rw-r--r-- | llama.h | 1 |
1 files changed, 1 insertions, 0 deletions
@@ -73,6 +73,7 @@ extern "C" { LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensors LLAMA_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4, // tok_embeddings.weight and output.weight are F16 LLAMA_FTYPE_MOSTLY_Q4_2 = 5, // except 1d tensors + LLAMA_FTYPE_MOSTLY_Q4_3 = 6, // except 1d tensors }; LLAMA_API struct llama_context_params llama_context_default_params(); |