ggml-org/llama.cpp

Pull Requests Commits

llama : reuse the tensor staging buffer and skip the direct-io bounce buffer

am17an committed 8 hours ago

a3185dea

codeowners : refresh teams and maintainer PRs link (#31316)

ggerganov committed 8 hours ago

Verified 604cc430

sycl: work-group argmax for short TOP_K rows (#31110)

newjordan committed 9 hours ago

Verified 39febd2a

cuda: fuse MoE (MUL_MAT_ID) gate/up and GLU in MMQ (#31188)

anujj committed 10 hours ago

Verified f5222f25

gguf-py: fix np_roundf dropping the sign (#31310)

ngxson committed 10 hours ago

Verified 28cb47c7

llama.cpp : bump version to 0.8.0 (#31309)

ggerganov committed 12 hours ago

Verified 8642541b

mtmd: enable video input support for LFM3 model family (#31283)

tdakhran committed 12 hours ago

Verified d4f1b18b

server : support tool calls in /v1/realtime (#31287)

tdakhran committed 12 hours ago

Verified d1eb0deb

ggerganov committed 13 hours ago

5d7793d6

ggml : bump version to v.28.0 (ggml/1676)

ggerganov committed 13 hours ago

ec941ee8

ui: build the reasoning menu from the template effort levels (#31273)

ServeurpersoCom committed 14 hours ago

Verified e486a4b5

tests : adjust constants to reduce irrelevant warnings (#31254)

ggerganov committed 14 hours ago

Verified f1a7dea9

cuda : add top-k MoE fusion for 384 experts (#31288)

developer0hye committed 15 hours ago

Verified a7af149a

vulkan: clamp A prefetch to end_k in coopmat1 int8 matmul (#30283)

SelfRef committed 15 hours ago

Verified 2864937a

chat : stop qwen3-coder reasoning and content before template separators (#30557)

7konigsberg committed 15 hours ago

Verified a71c00f0

ui : show save and reset settings buttons only when relevant (#31248)

allozaur committed 16 hours ago

Verified f4d8af60

ui: Support OGG audio attachments by converting them to WAV (#31237)

allozaur committed 16 hours ago

Verified 02d502f3

ui: Fix draft for new message disappearing after using "Add System Message" action (#31214)

allozaur committed 16 hours ago

Verified b1756b23

ui : Search dialog (#31173)

allozaur committed 16 hours ago

Verified 6886628c

server : add realtime conversation and transcription on /v1/realtime (#31269)

tdakhran committed 23 hours ago

Verified 8ca4060b

model : add LiquidAI/d1-2B and LiquidAI/d1-omni-2B (#31275)

tdakhran committed 1 day ago

Verified d05706b4

server: expose the reasoning effort levels of the chat template (#31271)

ServeurpersoCom committed 1 day ago

Verified 1b4554ef

tests: fix CI riscv timeout in test-llama-context (#31265)

ServeurpersoCom committed 1 day ago

Verified a9191799

HIP/CUDA: remove pragma unroll from loops in conv2d where itteration count cant be infered at compile time (#31255)

IMbackK committed 1 day ago

Verified 338fce65

model : add LiquidAI/LFM3-Audio-2B and LiquidAI/LFM3-Omni-2B (#31257)

tdakhran committed 1 day ago

Verified 19ff817f

common : support deferred tools and tool_reference content parts (#31256)

tdakhran committed 1 day ago

Verified eb7818c0

snapdragon: switch QDC artifact upload and log download to v2 API (#31240)

mengshengwu committed 1 day ago

Verified 6c445910

model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B (#31252)

tdakhran committed 1 day ago

Verified 19cce691

cmake: WASI doesn't support spawning (#30751)

MendyBerger committed 1 day ago

Verified e86f5fd1

llama : build the k-pool sparse selection independently of the kq_mask shape (#31243)

am17an committed 1 day ago

Verified 27b7a8a0

Older