ngxson/llama.cpp

Pull Requests Commits
server: fix data race in to_json_anthropic (#18283)
ngxson committed 85 days ago
Verified 3997c78e
release: update release workflow to store XCFramework as Zip file (#18284)
mattt committed 85 days ago
Verified ee746429
convert: rework ftype heuristics (#18214)
taronaeo committed 85 days ago
Verified a2831048
server: (docs) remove mention about extra_args (#18262)
ngxson committed 85 days ago
Verified 86af8481
tool/ex/tests: consistently free ctx, then model (#18168)
JohannesGaessler committed 85 days ago
Verified 147a5216
vulkan: Implement set_tensor_async and the event interfaces (#18047)
jeffbolznv committed 85 days ago
Verified e1f15b45
llama: fix RPC for -fit on (#18233)
JohannesGaessler committed 85 days ago
Verified 0e1ccf15
move copilot instructions to AGENTS.md (#18259)
ngxson committed 85 days ago
Verified 5e25ddeb
vulkan: fix im2col overflowing maxworkgroupcount (#18180)
jeffbolznv committed 86 days ago
Verified fd05c51c
vulkan/cuda: fix topk_moe with exp_probs_b (#18071)
jeffbolznv committed 86 days ago
Verified b365c3ff
vulkan: support GGML_UNARY_OP_XIELU (#18062)
jeffbolznv committed 86 days ago
Verified cb64222b
vulkan: in graph_optimize, try to group ADD operations (#18060)
jeffbolznv committed 86 days ago
Verified 6eb70818
Vulkan: some improvement on mul_mat_iq2_xs (#18031)
lovedheart committed 86 days ago
Verified 4117ae55
docs : fix links in parsing.md (#18245)
danbev committed 86 days ago
Verified 65e96a24
common : reorganize includes to prioritize vendored deps (#18222)
aldehir committed 86 days ago
Verified 9496bbb8
server: add auto-sleep after N seconds of idle (#18228)
ngxson committed 86 days ago
Verified ddcb75dd
tests: Avoid floating point precision false positives in SUM (#17471)
jeffbolznv committed 86 days ago
Verified 52ab19df
test-backend-ops: improve msvc build time (#18209)
jeffbolznv committed 86 days ago
Verified 5182dd64
Added comments explaining thread block size selection logic based on row count and column size, derived from historical commit context (#18212)
Aadeshveer committed 87 days ago
Verified 10b4f82d
server : [easy] fix per round speculative decode logging (#18211)
okuvshynov committed 87 days ago
Verified 408616ad
server: support load model on startup, support preset-only options (#18206)
ngxson committed 87 days ago
Verified 9e39a1e6
ci : remove non-windows zip artifacts (#18201)
CISC committed 87 days ago
Verified 74e05131
ci : only save ccache on master (#18207)
CISC committed 87 days ago
Verified f74747d8
ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations (#17977)
ngdxzy committed 87 days ago
Verified ce734a8a
arg: fix order to use short form before long form (#18196)
ServeurpersoCom committed 87 days ago
Verified 14931a82
llama : Changing off_t to size_t for Windows (#18204)
JTischbein committed 87 days ago
Verified f99ef53d
server: friendlier error msg when ctx < input (#18174)
am17an committed 88 days ago
Verified cc0a0434
presets: refactor, allow cascade presets from different sources, add global section (#18169)
ngxson committed 88 days ago
Verified 98c1c7a7
webui: Add editing attachments in user messages (#18147)
allozaur committed 88 days ago
Verified acb73d83
model-conversion : add verbose flag in run-org-model.py (#18194)
danbev committed 88 days ago
Verified 0a271d82
Terms Of Use Privacy Policy Refund Policy Impressum