ggerganov/llama.cpp

Pull Requests Commits

ui : keep llama.cpp servers as the only providers

allozaur committed 6 days ago

38f85919

ui : keep the table status working while a provider is active

allozaur committed 6 days ago

27973724

ui : offer the compat providers a selection instead of a load

allozaur committed 6 days ago

584d14ef

ui : hide the chat template on the remote providers

allozaur committed 6 days ago

c747f6d7

ui : show the selected model of the compat backends

allozaur committed 6 days ago

89647bda

ui : keep the selector list open on provider backends

allozaur committed 6 days ago

b0c4184b

test : seed the picker drawer through the providers data layer

allozaur committed 6 days ago

dd7cffa2

ui : resolve the layout probes through the providers api url

allozaur committed 6 days ago

934b95f8

ui : show every provider's models in the manager

allozaur committed 6 days ago

7891e7a6

ui : add the manage providers view

allozaur committed 6 days ago

e263e6e1

ui : route chat through the active provider

allozaur committed 6 days ago

a76734ba

ui : add the providers data layer

allozaur committed 6 days ago

118649c7

metal : fix moe_reduce fusion race when dst aliases its inputs (#30795)

ggerganov committed 6 days ago

Verified 098ce526

cann : refactor `#ifdef` blocks to avoid unreachable code after `return` (#25273)

rauletorresc committed 6 days ago

Verified f3fb2fcd

tests: add big-endian conversion tests (#30705)

taronaeo committed 6 days ago

Verified fda00f18

CUDA: fix warning for nonexistent extra semicolon (#30774)

JohannesGaessler committed 6 days ago

Verified 4866db35

ggml-cpu: vectorize ggml_v_silu and ggml_v_expf (#30791)

taronaeo committed 6 days ago

Verified 42b70032

cuda: cap MMQ Stream-K blocks at available work (#30553)

joelagnel committed 6 days ago

Verified 4c1004c4

qwen3 : bidirectional attention and dense_2 output projection (#30607)

fzowl committed 6 days ago

Verified 412a363c

ui: Fix markdown tables and lists being cut off on phones (#30687)

allozaur committed 6 days ago

Verified 2b57148b

tools : reject truncated GGUF tensor data in split and export tools (#30497)

ppAlexa committed 6 days ago

Verified 8f97d82d

tests : update Metal fusion baseline for beam (#30764)

ggerganov committed 6 days ago

Verified 27af5816

pi : turn .pi/gg into a system prompt addendum (#30762)

ggerganov committed 6 days ago

Verified 7db0d3d7

CUDA: fix shared-memory WAR hazard in flash_attn_ext_vec (#30573)

invaliddayta committed 6 days ago

Verified 99db719b

hexagon: FLASH_ATTN_EXT improvements for longer context and SWA models (#30731)

max-krasnyansky committed 6 days ago

Verified 31288c9c

hexagon: allow multistream FA (#30218)

jhen0409 committed 6 days ago

Verified fe8841a7

cuda : fuse qwen4exp's SCALE -> SIGMOID -> SCALE -> hc_post chain (#30602)

am17an committed 6 days ago

Verified ff5a2337

ggml: support non-contiguous SUM (CPU, CUDA, Vulkan, WebGPU) (#28851)

rssndr committed 6 days ago

Verified 0878f7c3

ggml-cuda: fattn do not check oob if k_VKQ_max = (k_VKQ_max - nbatch_fa) (#30572)

thelittlefireman committed 7 days ago

Verified 5aeac37a

Restore ESIMD kernels on FFN path (#30239)

malsbat committed 7 days ago

Verified 547208dc

Older