vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml : add Q5_0 and Q5_1 quantization (#1187)

Browse source 
* ggml : add Q5_0 quantization (cuBLAS only)

* ggml : fix Q5_0 qh -> uint32_t

* ggml : fix q5_0 histogram stats

* ggml : q5_0 scalar dot product

* ggml : q5_0 ARM NEON dot

* ggml : q5_0 more efficient ARM NEON using uint64_t masks

* ggml : rename Q5_0 -> Q5_1

* ggml : adding Q5_0 mode

* quantize : add Q5_0 and Q5_1 to map

* ggml : AVX2 optimizations for Q5_0, Q5_1 (#1195)

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
This commit is contained in:
Georgi Gerganov 2023-04-26 23:14:13 +03:00 • committed by GitHub
parent 87a6f846d3
commit 574406dc7e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
8 changed files with 711 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama.h
View file

@ -75,6 +75,8 @@ extern "C" {
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_FTYPE_MOSTLY_Q5_0 = 8, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q5_1 = 9, // except 1d tensors
};
LLAMA_API struct llama_context_params llama_context_default_params();