 Georgi Gerganov
|
3e63a58ef7
kv-cache : refactor the update/defrag mechanism (#13988)
|
il y a 7 mois |
|
Georgi Gerganov
|
0fc16b42e8
kv-cache : split implementation in separate sources (#13920)
|
il y a 7 mois |
|
Georgi Gerganov
|
3600cc2886
llama : use n_swa + n_ubatch cells for SWA cache (#13833)
|
il y a 8 mois |
|
Georgi Gerganov
|
12d0188c0d
kv-cache : refactor + add llama_memory_state_i (#13746)
|
il y a 8 mois |
|
Georgi Gerganov
|
81713121ee
kv-cells : track min/max used cells and per-sequence positions (#13808)
|
il y a 8 mois |
|
Georgi Gerganov
|
de2ef53a4b
kv-cache : rework kv_cell (#13706)
|
il y a 8 mois |
|
Georgi Gerganov
|
797f2ac062
kv-cache : simplify the interface (#13660)
|
il y a 8 mois |
|
Georgi Gerganov
|
a4090d1174
llama : remove llama_kv_cache_view API + remove deprecated (#13653)
|
il y a 8 mois |
|
Georgi Gerganov
|
e298d2fbd0
kv-cache : add SWA support (#13194)
|
il y a 8 mois |
|
Georgi Gerganov
|
e3a9421b78
kv-cache : fix out-of-bounds view during reserve graph (#13547)
|
il y a 8 mois |
|
Georgi Gerganov
|
c642bc014c
kv-cache : separate recurrent vs non-recurrent impl (#12799)
|
il y a 8 mois |
|
Georgi Gerganov
|
3e1d29348b
kv-cache : simplify + fix warning for recurrent models (#12756)
|
il y a 9 mois |
|
Georgi Gerganov
|
a10b36c91a
llama : refactor kv cache guard (#12695)
|
il y a 9 mois |
|
Georgi Gerganov
|
e0dbec0bc6
llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)
|
il y a 10 mois |
 mgroeber9110
|
5bbe6a9fe9
ggml : portability fixes for VS 2017 (#12150)
|
il y a 10 mois |
 Daniel Bevenius
|
3e69319772
llama : update llama_decode_internal ref [no ci] (#11840)
|
il y a 11 mois |
|
Georgi Gerganov
|
f66f582927
llama : refactor `src/llama.cpp` (#10902)
|
il y a 1 an |
