ngxson/llama.cpp

Pull Requests Commits

ngxson committed 7 hours ago

d4a3c4d4

correct gating func

ngxson committed 7 hours ago

3bf8f23c

somewhat working

ngxson committed 7 hours ago

55584783

ngxson committed 8 hours ago

237dfadf

add swa and per-layer n_head_kv

ngxson committed 9 hours ago

db8afa6a

add_sliding_window_pattern

ngxson committed 11 hours ago

09d3df9b

ngxson committed 11 hours ago

4e810162

runnable not incorrect

ngxson committed 12 hours ago

85937f90

ngxson committed 12 hours ago

86e8d1f1

rename mimov2 --> mimo2

ngxson committed 12 hours ago

bd806505

mimov2: convert ok

ngxson committed 13 hours ago

eb2cee1b

webui: apply webui_settings on first load (#18223)

ServeurpersoCom committed 15 hours ago

Verified 5b6c9bc0

server: fix crash with model not having BOS/EOS (#18321)

ngxson committed 16 hours ago

Verified 849d0211

model-conversion : add device option to run-org-model.py (#18318)

danbev committed 16 hours ago

Verified 8e3ead6e

rpc : add check for rpc buffer type (#18242)

struct committed 20 hours ago

Verified 12ee1763

ggml-hexagon: create generalized functions for cpu side op (#17500)

chraac committed 22 hours ago

Verified ed759777

model-conversion : add trust_remote_code for embedding scripts (#18288)

danbev committed 23 hours ago

Verified 847c35f7

[SYCL] replace llama-cli by llama-completion to rm the impact to test script (#18290)

arthw committed 1 day ago

Verified a6a552e4

model : fix div-by-zero for Nemotron V2 (#18309)

Alessandro98-git committed 1 day ago

Verified 96e33a81

model : Granite Embedding support (#15641)

ryan-mangeno committed 1 day ago

Verified dfc959b8

gguf-py : do not align the data start offset (#18291)

compilade committed 1 day ago

Verified 8f488073

ggml-hexagon: gelu optimization (#18151)

joeldushouyu committed 1 day ago

Verified bf6bc3c1

gen-docs: automatically update markdown file (#18294)

ngxson committed 1 day ago

Verified 179fd82a

llamafile: add rvv support for sgemm kernels (#18199)

taimur-10x committed 1 day ago

Verified d34d5ca1

opencl: unpack q4_0 for adreno in get_tensor (#18278)

lhez committed 1 day ago

Verified eb492bf4

vulkan: Extend rope fusions to allow mrope (#18264)

jeffbolznv committed 1 day ago

Verified e3b35ddf

server: prevent data race from HTTP threads (#18263)

ngxson committed 1 day ago

Verified 6ce863c8

server: fix data race in to_json_anthropic (#18283)

ngxson committed 1 day ago

Verified 3997c78e

release: update release workflow to store XCFramework as Zip file (#18284)

mattt committed 1 day ago

Verified ee746429

convert: rework ftype heuristics (#18214)

taronaeo committed 1 day ago

Verified a2831048

Older