 Georgi Gerganov
|
bcc0eb4591
llama : per-layer KV cache + quantum K cache (#4309)
|
2 gadi atpakaļ |
 cebtenzzre
|
b12fa0d1c1
build : link against build info instead of compiling against it (#3879)
|
2 gadi atpakaļ |
 slaren
|
16bc66d947
llama.cpp : split llama_context_params into model and context params (#3301)
|
2 gadi atpakaļ |
|
Cebtenzzre
|
8781013ef6
make : restore build-info.h dependency for several targets (#3205)
|
2 gadi atpakaļ |
|
Cebtenzzre
|
e6616cf0db
examples : add compiler version and target to build info (#2998)
|
2 gadi atpakaļ |
|
Cebtenzzre
|
3aefaab9e5
check C++ code with -Wmissing-declarations (#3184)
|
2 gadi atpakaļ |
|
Cebtenzzre
|
00d62adb79
fix some warnings from gcc and clang-tidy (#3038)
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
6381d4e110
gguf : new file format with flexible meta data (beta) (#2398)
|
2 gadi atpakaļ |
 Stephan Walter
|
1b107b8550
ggml : generalize `quantize_fns` for simpler FP16 handling (#1237)
|
2 gadi atpakaļ |
 Didzis Gosko
|
527b6fba1d
llama : make model stateless and context stateful (llama_state) (#1797)
|
2 gadi atpakaļ |
 Borislav Stanimirov
|
9cbf50c041
build : fix and ignore MSVC warnings (#1889)
|
2 gadi atpakaļ |
 Kawrakow
|
99009e72f8
ggml : add SOTA 2,3,4,5,6 bit k-quantizations (#1684)
|
2 gadi atpakaļ |
|
Stephan Walter
|
dc271c52ed
Remove unused n_parts parameter (#1509)
|
2 gadi atpakaļ |
 DannyDaemonic
|
f4cef87edf
Add git-based build information for better issue tracking (#1232)
|
2 gadi atpakaļ |
|
Kawrakow
|
38de86a711
llama : multi-threaded quantization (#1075)
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
eb17a026fd
quantize-stats : fix bug in --type argument
|
2 gadi atpakaļ |
 Pavol Rusnak
|
c56b715269
Expose type name from ggml (#970)
|
2 gadi atpakaļ |
|
Georgi Gerganov
|
9190e8eac8
llama : merge llama_internal.h into llama.h
|
2 gadi atpakaļ |
 comex
|
f963b63afa
Rewrite loading code to try to satisfy everyone:
|
2 gadi atpakaļ |
 unbounded
|
62cfc54f77
Add quantize-stats command for testing quantization (#728)
|
2 gadi atpakaļ |
