Georgi Gerganov
|
f9cd68398b
sampling : make sure samplers return at least 1 token (#13822)
|
7 maanden geleden |
DocShotgun
|
ffc727203a
sampling : make top_n_sigma no-op at <=0 or a single candidate (#13345)
|
8 maanden geleden |
oobabooga
|
91a86a6f35
sampling : don't consider -infinity values in top_n_sigma (#13344)
|
8 maanden geleden |
oobabooga
|
233461f812
sampling : Integrate Top-nσ into main sampling chain (and add it to the server) (#13264)
|
8 maanden geleden |
Georgi Gerganov
|
d9d398f84f
sampling : when top-k <= 0 -> noop (#13173)
|
8 maanden geleden |
Johannes Gäßler
|
dd373dd3bf
llama: fix error on bad grammar (#12628)
|
9 maanden geleden |
Olivier Chafik
|
669912d9a5
`tool-call`: fix Qwen 2.5 Coder support, add micro benchmarks, support trigger patterns for lazy grammars (#12034)
|
10 maanden geleden |
Vinesh Janarthanan
|
27e8a23300
sampling: add Top-nσ sampler (#11223)
|
11 maanden geleden |
Christian Fillion
|
7ee953a64a
llama : add llama_sampler_init for safe usage of llama_sampler_free (#11727)
|
11 maanden geleden |
Olivier Chafik
|
8b576b6c55
Tool call support (generic + native for Llama, Functionary, Hermes, Mistral, Firefunction, DeepSeek) w/ lazy grammars (#9639)
|
11 maanden geleden |
Georgi Gerganov
|
afa8a9ec9b
llama : add `llama_vocab`, functions -> methods, naming (#11110)
|
1 jaar geleden |
Georgi Gerganov
|
727368c60f
llama : use LLAMA_TOKEN_NULL (#11062)
|
1 jaar geleden |
Georgi Gerganov
|
f66f582927
llama : refactor `src/llama.cpp` (#10902)
|
1 jaar geleden |
Georgi Gerganov
|
644fd71b44
sampling : refactor + optimize penalties sampler (#10803)
|
1 jaar geleden |
wwoodsTM
|
5107e8cea3
DRY: Fixes clone functionality (#10192)
|
1 jaar geleden |
Georgi Gerganov
|
8d8ff71536
llama : remove Tail-Free sampling (#10071)
|
1 jaar geleden |
wwoodsTM
|
ff252ea48e
llama : add DRY sampler (#9702)
|
1 jaar geleden |
Georgi Gerganov
|
55e47786e3
llama : default sampling changes + greedy update (#9897)
|
1 jaar geleden |
Georgi Gerganov
|
99bd4ac28c
llama : infill sampling handle very long tokens (#9924)
|
1 jaar geleden |
Georgi Gerganov
|
755a9b2bf0
llama : add infill sampler (#9896)
|
1 jaar geleden |
MaggotHATE
|
fbc98b748e
sampling : add XTC sampler (#9742)
|
1 jaar geleden |
Georgi Gerganov
|
b0f27361f3
sampling : avoid expensive softmax during greedy sampling (#9605)
|
1 jaar geleden |
Daniel Bevenius
|
6443ddd985
llama : use reserve/emplace_back in sampler_sample (#9534)
|
1 jaar geleden |
Georgi Gerganov
|
0abc6a2c25
llama : llama_perf + option to disable timings during decode (#9355)
|
1 jaar geleden |
Gilad S.
|
bd35cb0ae3
feat: remove a sampler from a chain (#9445)
|
1 jaar geleden |
slaren
|
49006c67b4
llama : move random seed generation to the samplers (#9398)
|
1 jaar geleden |
slaren
|
5fb5e24811
llama : minor sampling refactor (2) (#9386)
|
1 jaar geleden |
slaren
|
19f4a7b296
llama : refactor samplers internal implementation (#9370)
|
1 jaar geleden |
Georgi Gerganov
|
f12295b8a9
llama : fix empty ring buffer push (#9358)
|
1 jaar geleden |
Georgi Gerganov
|
df270ef745
llama : refactor sampling v2 (#9294)
|
1 jaar geleden |