 Piotr Wilkin
|
ee52fe36f3
Modify sanity check to handle hybrid models
|
3 сар өмнө |
 Xuan-Son Nguyen
|
8f8f2274ee
convert : add Llama4ForCausalLM (#16042)
|
4 сар өмнө |
 Sigbjørn Skjæret
|
b8e09f08b9
model : add grok-2 support (#15539)
|
4 сар өмнө |
 Jie Fu (傅杰)
|
4f658855fa
llama : support T5 models with unequal number of encoder-decoder layers (#15909)
|
4 сар өмнө |
 Georgi Gerganov
|
c610b6c11b
kv-cache : fix SWA checks + disable cacheless iSWA (#15811)
|
4 сар өмнө |
 Daniel Bevenius
|
fb15d649ed
llama : add support for EmbeddingGemma 300m (#15798)
|
4 сар өмнө |
|
Georgi Gerganov
|
b730706a49
kv-cache : support layer reuse (#15504)
|
5 сар өмнө |
|
Georgi Gerganov
|
fd1234cb46
llama : add gpt-oss (#15091)
|
5 сар өмнө |
 Sam
|
ef0144c087
model: support GLM 4.5 family of models (#14939)
|
5 сар өмнө |
 Dongliang Wei
|
6c6e397aff
model : add support for SmallThinker series (#14898)
|
6 сар өмнө |
 Gabriel Larson
|
4762ad7316
model : make rope_yarn_log_mul optional for deepseek2 (#14896)
|
6 сар өмнө |
|
Georgi Gerganov
|
225e7a1438
llama : add high-throughput mode (#14363)
|
6 сар өмнө |
|
Gabriel Larson
|
4a4f426944
model : add Kimi-K2 support (#14654)
|
6 сар өмнө |
 Tarek Dakhran
|
f5e96b368f
model : support LiquidAI LFM2 hybrid family (#14620)
|
6 сар өмнө |
 compilade
|
5d46babdc2
llama : initial Mamba-2 support (#9126)
|
7 сар өмнө |
|
Xuan-Son Nguyen
|
8846aace49
model : gemma3n text-only (#14400)
|
7 сар өмнө |
|
Georgi Gerganov
|
4c9fdfbe15
ubatch : new splitting logic (#14217)
|
7 сар өмнө |
 Gabe Goodhart
|
edc4a29eff
memory : Hybrid recurrent cache (#13979)
|
7 сар өмнө |
|
Sigbjørn Skjæret
|
6385b843a8
llama : add RobertaForSequenceClassification reranker support (#13875)
|
8 сар өмнө |
|
Georgi Gerganov
|
d13d0f6135
hparams : initialize arrays (#13728)
|
8 сар өмнө |
|
Xuan-Son Nguyen
|
8a2afb7520
llama : allow custom list of swa_layers (#13726)
|
8 сар өмнө |
|
Georgi Gerganov
|
8e186ef0e7
hparams : support models for which all layers use SWA (#13682)
|
8 сар өмнө |
|
Georgi Gerganov
|
e298d2fbd0
kv-cache : add SWA support (#13194)
|
8 сар өмнө |
 AT
|
5f5e39e1ba
model : Nomic Embed Text V2 with Mixture-of-Experts (MoE) architecture (#12466)
|
9 сар өмнө |
 Juk Armstrong
|
daa422881a
llama : DeepSeek V2/V3 MLA implementation (#12801)
|
9 сар өмнө |
|
Xuan-Son Nguyen
|
1466621e73
llama : Support llama 4 text-only (#12791)
|
9 сар өмнө |
 Molly Sophia
|
7dfad387e3
llama: Add support for RWKV v7 architecture (#12412)
|
10 сар өмнө |
|
Georgi Gerganov
|
081bee8c64
hparams : add SWA rope parameters (#12374)
|
10 сар өмнө |
|
Georgi Gerganov
|
84d5475541
llama : fix Gemma3 SWA KV cache shift (#12373)
|
10 сар өмнө |
|
Georgi Gerganov
|
afa8a9ec9b
llama : add `llama_vocab`, functions -> methods, naming (#11110)
|
1 жил өмнө |
