diff options
author | Georgi Gerganov <ggerganov@gmail.com> | 2023-04-25 23:40:51 +0300 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-04-25 23:40:51 +0300 |
commit | 7a32fcb3b29f4db8aed8a85dc58eb958fb118153 (patch) | |
tree | b363c851cead2b5d6efced33cc461e37d8ed6bf8 /llama.h | |
parent | dd0eabc049fb1efc631cab8eb0a646808d704e18 (diff) |
ggml : add Q8_0 quantization format (rename the old one to Q8_1) (ARM NEON) (#1179)
* ggml : add Q8_0 quantization format (rename the old one to Q8_1)
* tests : fix test-quantize-fns
* ggml : finalize Q8_0 implementation
* ggml : use q4_0_q8_0 and q4_2_q8_0
* ggml : fix Q8_0 dot product bug (ARM)
* ggml : Q8_0 unroll x2
* ggml : fix bug - using wrong block type
* ggml : extend quantize_fns_t with "vec_dot_type"
* ggml : fix Q8_0 to use 255 values out of 256
* ggml : fix assert using wrong QK4_2 instead of QK4_3
Diffstat (limited to 'llama.h')
-rw-r--r-- | llama.h | 1 |
1 files changed, 1 insertions, 0 deletions
@@ -74,6 +74,7 @@ extern "C" { 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_FTYPE_MOSTLY_Q8_0 = 7, // except 1d tensors }; LLAMA_API struct llama_context_params llama_context_default_params(); |