 Joan Fontanals
|
f5d7b268ec
llama : add jina v2 base code (#7596)
|
1 year ago |
 Georgi Gerganov
|
2b3389677a
ggml : refactor rope norm/neox (#7634)
|
1 year ago |
|
Georgi Gerganov
|
1442677f92
common : refactor cli arg parsing (#7675)
|
1 year ago |
|
Georgi Gerganov
|
554c247caf
ggml : remove OpenCL (#7735)
|
1 year ago |
|
Georgi Gerganov
|
0cd6bd3483
llama : remove beam search (#7736)
|
1 year ago |
 jaime-m-p
|
3b38d48609
Per token attributes (#7685)
|
1 year ago |
 Radoslav Gerganov
|
bde7cd3cd9
llama : offload to RPC in addition to other backends (#7640)
|
1 year ago |
 0cc4m
|
3d7ebf6312
Vulkan Mixture of Experts (MoE) support (#7628)
|
1 year ago |
 zhangkaihuo
|
6f28a333c1
llama : MiniCPM support tied embeddings (#7664)
|
1 year ago |
|
Georgi Gerganov
|
549279d804
llama : avoid double token-to-piece cache (#7654)
|
1 year ago |
 Johannes Gäßler
|
9b596417af
CUDA: quantized KV support for FA vec (#7527)
|
1 year ago |
|
Georgi Gerganov
|
5921b8f089
llama : cache llama_token_to_piece (#7587)
|
1 year ago |
|
Georgi Gerganov
|
fb76ec31a9
ggml : fix YARN + add tests + add asserts (#7617)
|
1 year ago |
|
jaime-m-p
|
02c1ecad07
Tokenizer WPM fixes (#7500)
|
1 year ago |
 Giuseppe Scrivano
|
5442939fcc
llama : support small Granite models (#7481)
|
1 year ago |
 fairydreaming
|
ee3dff6b8e
Add support for DeepseekV2ForCausalLM (#7519)
|
1 year ago |
|
Georgi Gerganov
|
8b99e2aa66
llama : handle unknown utf8 bytes (#7588)
|
1 year ago |
 Bartowski
|
c429b33beb
llama : add Smaug 70B support (#7402)
|
1 year ago |
 Justine Tunney
|
00c6390793
main : don't print special tokens with --grammar (#6923)
|
1 year ago |
 Masaya, Kato
|
faa0e6979a
ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
|
1 year ago |
|
fairydreaming
|
fbca2f27fc
Add support for ArcticForCausalLM (#7020)
|
1 year ago |
 Tristan Druyen
|
007489e895
Fix phi3 chat template confusion with zephyr (#7449)
|
1 year ago |
 Daniel Bevenius
|
3015851c5a
llama : add getters for n_threads/n_threads_batch (#7464)
|
1 year ago |
|
Georgi Gerganov
|
55ac3b7aea
ci : use Pythia models instead of OpenLlama (#7470)
|
1 year ago |
|
fairydreaming
|
9b82476ee9
Add missing inference support for GPTNeoXForCausalLM (Pythia and GPT-NeoX base models) (#7461)
|
1 year ago |
|
Georgi Gerganov
|
a61a94e543
llama : rename n_ctx -> cache.size, less confusing (#0)
|
1 year ago |
|
Georgi Gerganov
|
e84b71c2c6
ggml : drop support for QK_K=64 (#7473)
|
1 year ago |
 slaren
|
b18532a4ef
phi3 : duplicate rope factors in each layer (#7447)
|
1 year ago |
|
Justine Tunney
|
03d8900ebe
llama : add missing model type names (#7445)
|
1 year ago |
 liuwei-git
|
201cc11afa
llama : add phi3 128K model support (#7225)
|
1 year ago |
