 Georgi Gerganov
|
56a00f0a2f
common : default --hf-file to --model (#6234)
|
1 year ago |
 fraxy-v
|
92397d87a4
convert-llama2c-to-ggml : enable conversion of GQA models (#6237)
|
1 year ago |
 Kawrakow
|
1d0331c12a
quantize: options for output and token embedding tensors qtype (#6239)
|
1 year ago |
 Pierrick Hymbert
|
dba1af6129
llama_model_loader: support multiple split/shard GGUFs (#6187)
|
1 year ago |
 Minsoo Cheong
|
ee804f6223
ci: apply concurrency limit for github workflows (#6243)
|
1 year ago |
|
Georgi Gerganov
|
80bd33bc2c
common : add HF arg helpers (#6234)
|
1 year ago |
 Nexesenex
|
e80f06d2a1
llama : correction of the attn.v.weight quantization for IQ3_XS (#6209)
|
1 year ago |
 Olivier Chafik
|
f77a8ffd3b
tests : conditional python & node json schema tests (#6207)
|
1 year ago |
|
Olivier Chafik
|
72114edf06
json-schema-to-grammar : fix order of props + non-str const/enum (#6232)
|
1 year ago |
 slaren
|
2f0e81e053
cuda : add LLAMA_CUDA_NO_PEER_COPY to workaround broken ROCm p2p copy (#6208)
|
1 year ago |
 Xiaoyi Chen
|
29ab270e65
readme : add RecurseChat to the list of UIs (#6219)
|
1 year ago |
 Jan Boon
|
6b8bb3a31d
server : fix n_keep always showing as 0 in response (#6211)
|
1 year ago |
|
Georgi Gerganov
|
68e210b354
server : enable continuous batching by default (#6231)
|
1 year ago |
|
Georgi Gerganov
|
b3e94f26ba
metal : proper assert for mat-mat memory alignment (#6225)
|
1 year ago |
 Vaibhav Srivastav
|
b2075fd6a5
ci : add CURL flag for the mac builds (#6214)
|
1 year ago |
|
Georgi Gerganov
|
95d576b48e
metal : pad n_ctx by 32 (#6177)
|
1 year ago |
 Neo Zhang Jianyu
|
59c17f02de
add blog link (#6222)
|
1 year ago |
 DAN™
|
fa046eafbc
Fix params underscore convert to dash. (#6203)
|
1 year ago |
|
Jan Boon
|
be07a03217
server : update readme doc from `slot_id` to `id_slot` (#6213)
|
1 year ago |
|
slaren
|
d0a71233fb
cuda : disable host register by default (#6206)
|
1 year ago |
 semidark
|
f372c49ccd
Corrected typo to wrong file (#6199)
|
1 year ago |
|
Georgi Gerganov
|
924ce1dce7
tests : disable system() calls (#6198)
|
1 year ago |
|
slaren
|
03a8f8fafe
cuda : fix LLAMA_CUDA_F16 build (#6197)
|
1 year ago |
|
Kawrakow
|
cfd3be76e3
ggml : same IQ4_NL quantization for CPU/CUDA/Metal (#6196)
|
1 year ago |
|
Olivier Chafik
|
5b7b0ac8df
json-schema-to-grammar improvements (+ added to server) (#5978)
|
1 year ago |
|
Vaibhav Srivastav
|
1943c01981
ci : fix indentation error (#6195)
|
1 year ago |
|
Vaibhav Srivastav
|
5e43ba8742
build : add mac pre-build binaries (#6182)
|
1 year ago |
|
Kawrakow
|
76aa30a263
Add ability to use Q5_0, Q5_1, and IQ4_NL for quantized K cache (#6183)
|
1 year ago |
 AidanBeltonS
|
c5b8595e3f
Add nvidia and amd backends (#6157)
|
1 year ago |
|
slaren
|
42e21c6882
cuda : fix conflict with std::swap (#6186)
|
1 year ago |
