| .. |
|
CMakeLists.txt
|
4201deae9c
common: introduce http.h for httplib-based client (#16373)
|
преди 3 месеца |
|
arg.cpp
|
6f5d924637
common : Update the docs on -t --threads (#16236)
|
преди 3 месеца |
|
arg.h
|
364a7a6d4a
common : remove common_has_curl() (#16351)
|
преди 3 месеца |
|
base64.hpp
|
381efbf480
llava : expose as a shared library for downstream projects (#3613)
|
преди 2 години |
|
build-info.cpp.in
|
cc8d081879
cmake: Add ability to pass in LLAMA_BUILD_NUMBER/COMMIT (#14167)
|
преди 7 месеца |
|
chat-parser.cpp
|
2c301e91ab
common : handle unicode during partial json parsing (#16526)
|
преди 3 месеца |
|
chat-parser.h
|
34fcc5a4ac
model : Apertus model implementation (#15852)
|
преди 3 месеца |
|
chat.cpp
|
12bbc3fa50
refactor: centralize CoT parsing in backend for streaming mode (#16394)
|
преди 3 месеца |
|
chat.h
|
d00cbea63c
server : host-memory prompt caching (#16391)
|
преди 3 месеца |
|
common.cpp
|
3df2244df4
llama : add --no-host to disable host buffers (#16310)
|
преди 3 месеца |
|
common.h
|
4b2dae383d
common : update presets (#16504)
|
преди 3 месеца |
|
console.cpp
|
8277a817f1
console : utf-8 fix for windows stdin (#9690)
|
преди 1 година |
|
console.h
|
6381d4e110
gguf : new file format with flexible meta data (beta) (#2398)
|
преди 2 години |
|
http.h
|
4201deae9c
common: introduce http.h for httplib-based client (#16373)
|
преди 3 месеца |
|
json-partial.cpp
|
2c301e91ab
common : handle unicode during partial json parsing (#16526)
|
преди 3 месеца |
|
json-partial.h
|
53f925074d
sync : vendor (#13901)
|
преди 7 месеца |
|
json-schema-to-grammar.cpp
|
79967ec596
grammar : use int64_t to avoid int overflows in int schema to grammar conversion logic (#16626)
|
преди 3 месеца |
|
json-schema-to-grammar.h
|
53f925074d
sync : vendor (#13901)
|
преди 7 месеца |
|
llguidance.cpp
|
43dfd741a5
llguidance : set tokenizer slices to default (#13424)
|
преди 8 месеца |
|
log.cpp
|
408ff524b4
Implement --log-colors with always/never/auto (#15792)
|
преди 4 месеца |
|
log.h
|
408ff524b4
Implement --log-colors with always/never/auto (#15792)
|
преди 4 месеца |
|
ngram-cache.cpp
|
5bbe6a9fe9
ggml : portability fixes for VS 2017 (#12150)
|
преди 10 месеца |
|
ngram-cache.h
|
727368c60f
llama : use LLAMA_TOKEN_NULL (#11062)
|
преди 1 година |
|
regex-partial.cpp
|
3198405e98
`common`: add partial regex support (#12808)
|
преди 8 месеца |
|
regex-partial.h
|
3198405e98
`common`: add partial regex support (#12808)
|
преди 8 месеца |
|
sampling.cpp
|
e789095502
llama: print memory breakdown on exit (#15860)
|
преди 3 месеца |
|
sampling.h
|
e92d53b29e
sampling : optimize samplers by reusing bucket sort (#15665)
|
преди 4 месеца |
|
speculative.cpp
|
e92d53b29e
sampling : optimize samplers by reusing bucket sort (#15665)
|
преди 4 месеца |
|
speculative.h
|
94933c8c2e
server : implement universal assisted decoding (#12635)
|
преди 5 месеца |
Olivier Chafik
79967ec596
grammar : use int64_t to avoid int overflows in int schema to grammar conversion logic (#16626)