 Georgi Gerganov
|
4d98d9a656
sync : ggml (SD ops, tests, kernels) (#4444)
|
il y a 2 ans |
 Jared Van Bortel
|
70f806b821
build : detect host compiler and cuda compiler separately (#4414)
|
il y a 2 ans |
 Siwen Yu
|
9fb13f9584
common : add `--version` option to show build info in CLI (#4433)
|
il y a 2 ans |
|
Georgi Gerganov
|
113f9942fc
readme : update hot topics
|
il y a 2 ans |
 slaren
|
799a1cb13b
llama : add Mixtral support (#4406)
|
il y a 2 ans |
 kalomaze
|
fecac45658
server : tweak default sampling parameters (#4367)
|
il y a 2 ans |
 Richard Kiss
|
9494d7c477
english : use `typos` to fix comments and logs (#4354)
|
il y a 2 ans |
|
Jared Van Bortel
|
6138963fb2
build : target Windows 8 for standard mingw-w64 (#4405)
|
il y a 2 ans |
 crasm
|
6391817cd1
llama : document logits_all deprecation (#4418)
|
il y a 2 ans |
 Vladimir Zorin
|
d9d4cfef64
server : fix local model name in server (#4420)
|
il y a 2 ans |
 Taikono-Himazin
|
41a11aaf99
ggml : increased GGML_MAX_PARAMS to allow finetuning of 70b models (#4424)
|
il y a 2 ans |
 Yueh-Po Peng
|
8a7b2fa528
Update README.md (#4388)
|
il y a 2 ans |
 Xiang (Kevin) Li
|
e18f7345a3
grammar : revert the replacement of llama_token_to_piece with id_to_token (#4396)
|
il y a 2 ans |
|
Georgi Gerganov
|
fe680e3d10
sync : ggml (new ops, tests, backend, etc.) (#4359)
|
il y a 2 ans |
|
Georgi Gerganov
|
bcc0eb4591
llama : per-layer KV cache + quantum K cache (#4309)
|
il y a 2 ans |
 Hongyu Ouyang
|
81bc9214a3
train : fix #4227 (double free in examples/train-text-from-scratch/train-text-from-scratch.cpp) (#4351)
|
il y a 2 ans |
|
Georgi Gerganov
|
05cd6e5036
server : recognize cache_prompt parameter in OAI API (#4347)
|
il y a 2 ans |
|
Georgi Gerganov
|
caa9249217
common : fix compile warning
|
il y a 2 ans |
 stduhpf
|
da5eaef1f3
speculative : support `--color` (#4343)
|
il y a 2 ans |
 Marcus Dunn
|
5f6e0c0dff
grammar : pre-computed pieces + reserve mem + less string copies (#4330)
|
il y a 2 ans |
 Kerfuffle
|
5aa365d88f
llama : allow overriding GGUF metadata when loading model (#4092)
|
il y a 2 ans |
 MaggotHATE
|
52c8bc3cf3
sampling : custom samplers order (#4285)
|
il y a 2 ans |
 kchro3
|
e4b76bbe31
swift : revert compiler checks for swift package (#4332)
|
il y a 2 ans |
 Daniel Bevenius
|
23b5e12eb5
simple : update error message for KV cache check (#4324)
|
il y a 2 ans |
 Miwa / Ensan
|
d208995c6d
swift : fix concatenation method to avoid invalid UTF8 stringfication (#4325)
|
il y a 2 ans |
|
Miwa / Ensan
|
5c9f90cba1
swift : fix prompt tokenization logic (#4321)
|
il y a 2 ans |
 Ikko Eltociear Ashimine
|
4fa44e84ad
grammar-parser : fix typo (#4318)
|
il y a 2 ans |
|
Georgi Gerganov
|
fbbc42827b
ggml : reuse ggml_get_n_tasks() in ggml_graph_plan() (#4308)
|
il y a 2 ans |
|
Georgi Gerganov
|
adf3de4f69
ggml : fix soft max out-of-bounds access (#4307)
|
il y a 2 ans |
 Ed Lee
|
33e171d1e9
server : fix OpenAI API `stop` field to be optional (#4299)
|
il y a 2 ans |
