ggml-org--ggml

master

7840aaba19 · sync : llama.cpp · 最后更新于 2026-09-09 12:31:37 +00:00

fix-mul-mat c5a21cd21b · ggml : fix mul_mat src1 indexing when src1 is not contiguous · 最后更新于 2023-07-14 08:01:57 +00:00 frostbyte_neo	4273 1	ZIP TAR.GZ
gq 724c45d545 · ggml : finalize the Q4_1 quantization for ARM_NEON · 最后更新于 2023-03-29 19:14:30 +00:00 frostbyte_neo	4522 38	ZIP TAR.GZ
cerebras-wip 23cc0e4425 · wip wip wip · 最后更新于 2023-03-29 17:05:20 +00:00 frostbyte_neo	4522 1	ZIP TAR.GZ
llama 7f32376b70 · llama : initial working FP16 + 4-bit Q4_0 · 最后更新于 2023-03-10 18:01:56 +00:00 frostbyte_neo	4528 38	ZIP TAR.GZ
4bit baeb88b858 · tests : add 4-bit Clover-based quantization · 最后更新于 2023-02-20 22:33:35 +00:00 frostbyte_neo	4533 1	ZIP TAR.GZ
t5 1d38a69d7c · t5 : initial load in ggml · 最后更新于 2023-01-02 14:11:13 +00:00 frostbyte_neo	4551 3	ZIP TAR.GZ
experiments/blocking 3afb833f84 · wip : unsuccessful attempts speeding mul_mat using blocking · 最后更新于 2022-10-13 19:19:43 +00:00 frostbyte_neo	4566 1	ZIP TAR.GZ

分支列表