 Olivier Chafik
|
db288b60cb
`tool-call`: command r7b fix for normal responses (#11608)
|
пре 11 месеци |
 Shelby Jenkins
|
106045e7bb
readme : add llm_client Rust crate to readme bindings (#11628)
|
пре 11 месеци |
 Jhen-Jie Hong
|
f117d84b48
swift : fix llama-vocab api usage (#11645)
|
пре 11 месеци |
|
Jhen-Jie Hong
|
534c46b53c
metal : use residency set for other platforms (#11648)
|
пре 11 месеци |
 Georgi Gerganov
|
387a1598ca
authors : update
|
пре 11 месеци |
|
Georgi Gerganov
|
7c9e0ca520
sync : ggml
|
пре 11 месеци |
 Christian Kastner
|
8f8290ada9
cmake: Add ability to pass in GGML_BUILD_NUMBER (ggml/1096)
|
пре 11 месеци |
|
Georgi Gerganov
|
b34aedd558
ci : do not stale-close roadmap issues
|
пре 11 месеци |
|
Olivier Chafik
|
cde3833239
`tool-call`: allow `--chat-template chatml` w/ `--jinja`, default to chatml upon parsing issue, avoid double bos (#11616)
|
пре 11 месеци |
 Xuan-Son Nguyen
|
b3451785ac
server : (webui) revert hacky solution from #11626 (#11634)
|
пре 11 месеци |
 Woof Dog
|
1d1e6a90bc
server : (webui) allow typing and submitting during llm response (#11626)
|
пре 11 месеци |
 Daniel Bevenius
|
5598f475be
server : remove CPPHTTPLIB_NO_EXCEPTIONS define (#11622)
|
пре 11 месеци |
|
Georgi Gerganov
|
8ec05832fa
sync : ggml
|
пре 11 месеци |
 Johannes Gäßler
|
21c84b5d2d
CUDA: fix Volta FlashAttention logic (#11615)
|
пре 11 месеци |
 mashdragon
|
d92cb67e37
server : (webui) Fix Shift+Enter handling (#11609)
|
пре 11 месеци |
|
Johannes Gäßler
|
6eecde3cc8
HIP: fix flash_attn_stream_k_fixup warning (#11604)
|
пре 11 месеци |
 uvos
|
396856b400
CUDA/HIP: add support for selectable warp size to mmv (#11519)
|
пре 11 месеци |
|
uvos
|
4d0598e144
HIP: add GGML_CUDA_CC_IS_* for amd familys as increasing cc archtectures for amd gpus are not supersets of eatch other (#11601)
|
пре 11 месеци |
|
Olivier Chafik
|
90f9b88afb
nit: more informative crash when grammar sampler fails (#11593)
|
пре 11 месеци |
|
Johannes Gäßler
|
864a0b67a6
CUDA: use mma PTX instructions for FlashAttention (#11583)
|
пре 11 месеци |
 Eric Curtin
|
84ec8a58f7
Name colors (#11573)
|
пре 11 месеци |
|
Olivier Chafik
|
bfcce4d693
`tool-call`: support Command R7B (+ return tool_plan "thoughts" in API) (#11585)
|
пре 11 месеци |
|
Olivier Chafik
|
69804487e0
Fix exotic ci env that lacks ostringstream::str (#11581)
|
пре 11 месеци |
 Michał Moskal
|
ff227703d6
sampling : support for llguidance grammars (#10224)
|
пре 11 месеци |
 piDack
|
0cec062a63
llama : add support for GLM-Edge and GLM-Edge-V series models (#10573)
|
пре 11 месеци |
|
Olivier Chafik
|
53debe6f3c
ci: use sccache on windows HIP jobs (#11553)
|
пре 11 месеци |
|
Olivier Chafik
|
cfd74c86db
`sync`: minja (https://github.com/google/minja/commit/418a2364b56dc9be4ed9a1a2b0fb16fb53a7a22e) (#11574)
|
пре 11 месеци |
|
Eric Curtin
|
ecef206ccb
Implement s3:// protocol (#11511)
|
пре 11 месеци |
|
Olivier Chafik
|
5bbc7362cb
ci: simplify cmake build commands (#11548)
|
пре 11 месеци |
|
Olivier Chafik
|
aa6fb13213
`ci`: use sccache on windows instead of ccache (#11545)
|
пре 11 месеци |
