 Johannes Gäßler
|
95fb0aefab
CUDA: remove incorrect precision check (#7454)
|
1 year ago |
 Georgi Gerganov
|
3e5faa8503
cuda : fix rope + add tests (#7452)
|
1 year ago |
 liuwei-git
|
201cc11afa
llama : add phi3 128K model support (#7225)
|
1 year ago |
|
Georgi Gerganov
|
6369bf0433
metal : handle F16 inf values, fix FA partial offload (#7434)
|
1 year ago |
 Olivier Chafik
|
e402de364b
`grammars`: fix resampling logic regression (#7424)
|
1 year ago |
|
Johannes Gäßler
|
fcf6538ba6
CUDA: fix unused warning in mmq.cu (#7442)
|
1 year ago |
|
Georgi Gerganov
|
c3f8d58356
tests : test-tokenizer-0.sh print more info (#7402)
|
1 year ago |
 Amir
|
11474e756d
examples: cache hf model when --model not provided (#7353)
|
1 year ago |
|
Johannes Gäßler
|
d8ee902227
CUDA: deduplicate mmq code (#7397)
|
1 year ago |
 jaime-m-p
|
d7e852c1bc
Tokenizer SPM fixes for phi-3 and llama-spm (bugfix) (#7425)
|
1 year ago |
|
jaime-m-p
|
917dc8cfa6
Tokenizer SPM fixes for phi-3 and llama-spm (#7375)
|
1 year ago |
|
Georgi Gerganov
|
fabf30b4c4
llama : remove Persimmon (#7408)
|
1 year ago |
|
Johannes Gäßler
|
20385cebcc
perplexity: update README FP16 results [no ci] (#7413)
|
1 year ago |
 Radoslav Gerganov
|
db10f01310
rpc : track allocated buffers (#7411)
|
1 year ago |
|
Georgi Gerganov
|
3bc10cb485
server : fix temperature + disable some tests (#7409)
|
1 year ago |
 AidanBeltonS
|
6bf9b66fa3
[SYCL] Update SYCL upscale operation (#7321)
|
1 year ago |
 Bingan
|
26cd4237bc
Update README.md (#7410)
|
1 year ago |
 Herman Semenov
|
213e90ed73
ggml-opencl, llama: using reserve() if count already known (#7272)
|
1 year ago |
 junchao-loongson
|
65c58207ec
ggml : add loongarch lsx and lasx support (#6454)
|
1 year ago |
|
Georgi Gerganov
|
1cc0155d04
server : tuning tests (#7388)
|
1 year ago |
|
Georgi Gerganov
|
e932094d58
server : return error on too large embedding input (#7389)
|
1 year ago |
|
Georgi Gerganov
|
2789baf480
tests : fix --keep_split -> --keep-split (#7374)
|
1 year ago |
 Srihari-mcw
|
33c8d50acc
Add provisions for windows support for BF16 code including CMake provision for enabling AVX512_BF16 (#7258)
|
1 year ago |
 slaren
|
d359f30921
llama : remove MPI backend (#7395)
|
1 year ago |
 Fred Douglas
|
1ea2a0036e
quantize : fix --keep-split check (#7374)
|
1 year ago |
 0cc4m
|
f030ec1f7a
Vulkan Embedding Fix (#7360)
|
1 year ago |
|
slaren
|
e4e6f67be6
ggml : fix another case of quants nans (#7387)
|
1 year ago |
|
Johannes Gäßler
|
5ca49cbecd
ggml: implement quantized KV cache for FA (#7372)
|
1 year ago |
|
Johannes Gäßler
|
1b01f06db0
server: add test for token probs (#7347)
|
1 year ago |
|
Johannes Gäßler
|
41858392e1
server: fix seed being reported back (#7382)
|
1 year ago |
