 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 |
|
Olivier Chafik
|
a83f528688
`tool-call`: fix llama 3.x and functionary 3.2, play nice w/ pydantic_ai package, update readme (#11539)
|
пре 11 месеци |
|
Olivier Chafik
|
b1bcd309fc
fix stop regression (#11543)
|
пре 11 месеци |
|
Olivier Chafik
|
5783575c9d
Fix chatml fallback for unsupported builtin templates (when --jinja not enabled) (#11533)
|
пре 11 месеци |
|
Olivier Chafik
|
4a2b196d03
server : fix --jinja when there's no tools or schema (typo was forcing JSON) (#11531)
|
пре 11 месеци |
 Steve Grubb
|
1bd3047a93
common: Add missing va_end (#11529)
|
пре 11 месеци |
 Daniel Bevenius
|
a2df2787b3
server : update help metrics processing/deferred (#11512)
|
пре 11 месеци |
|
Olivier Chafik
|
553f1e46e9
`ci`: ccache for all github worfklows (#11516)
|
пре 11 месеци |
|
Olivier Chafik
|
8b576b6c55
Tool call support (generic + native for Llama, Functionary, Hermes, Mistral, Firefunction, DeepSeek) w/ lazy grammars (#9639)
|
пре 11 месеци |
 uvos
|
27d135c970
HIP: require at least HIP 5.5
|
пре 11 месеци |
|
uvos
|
6af1ca48cb
HIP: Prepare reduction operators for wave 64
|
пре 11 месеци |
|
uvos
|
c300e68ef4
CUDA/HIP: add warp_size to cuda_device_info
|
пре 11 месеци |
|
Olivier Chafik
|
3d804dec76
sync: minja (#11499)
|
пре 11 месеци |
 mgroeber9110
|
ffd0821c57
vocab : correctly identify LF token for GPT-2 style BPE tokenizer (#11496)
|
пре 11 месеци |
|
Daniel Bevenius
|
4314e56c4f
server : use lambda instead of std::bind (#11507)
|
пре 11 месеци |
 Isaac McFadyen
|
496e5bf46b
server : (docs) added response format for /apply-template [no ci] (#11503)
|
пре 11 месеци |
 Guspan Tanadi
|
7919256c57
readme : reference examples relative links (#11505)
|
пре 11 месеци |
|
Daniel Bevenius
|
e0449763a4
server : update json snippets in README.md [no ci] (#11492)
|
пре 11 месеци |
 Nigel Bosch
|
eb7cf15a80
server : add /apply-template endpoint for additional use cases of Minja functionality (#11489)
|
пре 11 месеци |
 Rémy Oudompheng
|
66ee4f297c
vulkan: implement initial support for IQ2 and IQ3 quantizations (#11360)
|
пре 11 месеци |
|
Daniel Bevenius
|
e51c47b401
server : update auto gen files comments [no ci] (#11484)
|
пре 11 месеци |
 Jeff Bolz
|
2711d0215f
vulkan: Catch pipeline creation failure and print an error message (#11436)
|
пре 11 месеци |
 Eric Curtin
|
f0d4b29edf
Parse https://ollama.com/library/ syntax (#11480)
|
пре 11 месеци |
 Georgi Gerganov
|
815857791d
sync : ggml
|
пре 11 месеци |
 William Tambellini
|
1a0e87d291
ggml : add option to not print stack on abort (ggml/1081)
|
пре 11 месеци |
 issixx
|
d2e518e9b4
ggml-cpu : fix ggml_graph_compute_thread did not terminate on abort. (ggml/1065)
|
пре 1 година |
|
Daniel Bevenius
|
b636228c0a
embedding : enable --no-warmup option (#11475)
|
пре 11 месеци |
 Molly Sophia
|
325afb370a
llama: fix missing k_cache store for rwkv6qwen2 (#11445)
|
пре 11 месеци |
 Emreerdog
|
794fe23f29
cmake: add hints for locating ggml on Windows using Llama find-package (#11466)
|
пре 11 месеци |
