 Georgi Gerganov
|
812939a9e9
model : more uniform output id handling (#14275)
|
hai 7 meses |
 Gabe Goodhart
|
edc4a29eff
memory : Hybrid recurrent cache (#13979)
|
hai 7 meses |
 Đinh Trọng Huy
|
ad590be98c
model : add NeoBERT (#14164)
|
hai 7 meses |
 Bartowski
|
d7da8dc83a
model : Add support for Arcee AI's upcoming AFM model (#14185)
|
hai 7 meses |
 Mikko Juola
|
9ae4143bc6
model : add dots.llm1 architecture support (#14044) (#14118)
|
hai 7 meses |
 compilade
|
dad5c44398
kv-cache : avoid modifying recurrent cells when setting inputs (#13834)
|
hai 7 meses |
 Sigbjørn Skjæret
|
3678b838bb
llama : support GEGLU for jina-bert-v2 (#14090)
|
hai 7 meses |
|
Sigbjørn Skjæret
|
0974ad7a7c
llama : fix llama_model_chat_template with template name (LLM_KV with suffix) (#14050)
|
hai 7 meses |
|
Sigbjørn Skjæret
|
d17a809ef0
llama : support multiple classifier outputs and labels (#13940)
|
hai 7 meses |
|
Georgi Gerganov
|
5582c49c39
gemma : more consistent attention scaling for v2 and v3 (#13951)
|
hai 7 meses |
|
Georgi Gerganov
|
0fc16b42e8
kv-cache : split implementation in separate sources (#13920)
|
hai 7 meses |
|
Georgi Gerganov
|
3600cc2886
llama : use n_swa + n_ubatch cells for SWA cache (#13833)
|
hai 7 meses |
|
Georgi Gerganov
|
12d0188c0d
kv-cache : refactor + add llama_memory_state_i (#13746)
|
hai 7 meses |
|
Đinh Trọng Huy
|
291f2b6913
llama : add support for DistilBert (#13907)
|
hai 7 meses |
 zhangkaihuo
|
2c90da4c7e
llama : use llm_build_granite for minicpm (#13911)
|
hai 7 meses |
|
Sigbjørn Skjæret
|
e83ba3e460
llama : add support for jina-reranker-v2 (#13900)
|
hai 7 meses |
|
Sigbjørn Skjæret
|
6385b843a8
llama : add RobertaForSequenceClassification reranker support (#13875)
|
hai 7 meses |
 Piotr Jasiukajtis
|
4032ca4066
llama : add support for Qwen3 MoE tied word embeddings (#13768)
|
hai 8 meses |
|
Georgi Gerganov
|
d13d0f6135
hparams : initialize arrays (#13728)
|
hai 8 meses |
 Xuan-Son Nguyen
|
8a2afb7520
llama : allow custom list of swa_layers (#13726)
|
hai 8 meses |
|
Georgi Gerganov
|
8a1d206f1d
tts : fix n_ubatch + make WavTokenizer cache-less (#13713)
|
hai 8 meses |
|
Georgi Gerganov
|
797f2ac062
kv-cache : simplify the interface (#13660)
|
hai 8 meses |
|
Georgi Gerganov
|
b44890df2e
model : disable SWA for Phi models (#13676)
|
hai 8 meses |
|
Georgi Gerganov
|
be0239693c
model : fix llama4 graph (#13663)
|
hai 8 meses |
|
Georgi Gerganov
|
e298d2fbd0
kv-cache : add SWA support (#13194)
|
hai 8 meses |
 Gabe Goodhart
|
5e7d95e22e
fix: Move build_inp_pos to the top of the graph section for build_granite (#13538)
|
hai 8 meses |
|
Gabe Goodhart
|
d590cd4c24
model : Granite MoE shared (#13269)
|
hai 8 meses |
 Johannes Gäßler
|
10d2af0eaa
llama/ggml: add LLM training support (#10544)
|
hai 8 meses |
 Diego Devesa
|
27ebfcacba
llama : do not crash if there is no CPU backend (#13395)
|
hai 8 meses |
|
Xuan-Son Nguyen
|
3f96aeff39
llama : one-off chat template fix for Mistral-Small-2503 (#13398)
|
hai 8 meses |
