ggerganov/llama.cpp

Pull Requests Commits

llama_moe_cache: run everything without syncs

am17an committed 1 hour ago

a105865f

chat : fix Muse Spark ignoring response_format json_schema (#32017)

rmz-24 committed 12 hours ago

Verified aa75ff32

ui: keep a hidden sidebar from intercepting clicks again (#32035)

allozaur committed 13 hours ago

Verified b5cb8f6f

ui: update the README banner to the new llama.cpp cover and remove the legacy brand assets (#31789)

allozaur committed 13 hours ago

Verified 85822a7a

ui: show audio in Discover for models whose mmproj has an audio encoder (#31837)

allozaur committed 13 hours ago

Verified 2b66bfbf

ci : bump ty to 0.0.100 (#32030)

CISC committed 14 hours ago

Verified 582eab71

mtmd: fix string format, rename to mtmd_format (#32040)

ngxson committed 15 hours ago

Verified 7c2653ac

ui: move Models, MCP and Settings to a sidebar footer (#32022)

allozaur committed 17 hours ago

Verified 7592b61d

cuda : remove unused ids and type_src0 parameters (#32018)

danbev committed 17 hours ago

Verified cd4ed902

llama : Add FP8 KV-cache support (#28392)

ORippler committed 18 hours ago

Verified 202c1a22

Merge pull request #32015 from ggml-org/gg/fix-sampling-output-aliasing

ggerganov committed 19 hours ago

Verified e24d0b4c

llama : reuse backend scheduler when sched_params are compatible (#32009)

ggerganov committed 19 hours ago

Verified 3e86ce2c

ggml : add MUL_MAT src1 precision tests and gating (#31896)

ggerganov committed 20 hours ago

Verified 494d0f47

ui: gather chats into named groups in the sidebar (#32012)

allozaur committed 20 hours ago

Verified 281e7dad

ui: show settings save button even while focused (#32010)

hmblair committed 21 hours ago

Verified 90e3841d

cuda : remove unused quantize_cuda_t typedef (#31884)

danbev committed 21 hours ago

Verified c75fba30

graph: mark every sampling chain as an output to prevent aliasing

ggerganov committed 22 hours ago

40f4f4ac

ui: expand the chat form into a large editor for longer messages (#31973)

allozaur committed 22 hours ago

Verified 54cde710

convert : mark Sarvam byte fallback tokens as BYTE (#32006)

rmz-24 committed 22 hours ago

Verified 7d6fd845

convert : workaround conversion for p-e-w/gpt-oss-20b-heretic-ara-v4 (#31860)

linellazatin committed 23 hours ago

Verified 4ed88da2

cuda : halve VMM pool reservation size and retry on cuMemAddressReserve failure (#30542)

mike10010100 committed 23 hours ago

Verified ac931b5b

CUDA: skip graph compat check when graph is being reused (#32004)

am17an committed 1 day ago

Verified 15759e0c

mtmd: share compute buffer among subcontext (#31945)

ngxson committed 1 day ago

Verified a591cbe1

server: fix output lang for realtime api (#31982)

ngxson committed 1 day ago

Verified 203b5f40

mtmd: add kimi K3.1 vision (#31980)

ngxson committed 1 day ago

Verified 351025ec

ggml : fix FreeBSD Vulkan build by setting _XOPEN_SOURCE=700 (#31701)

angt committed 1 day ago

Verified 0d9eacad

CUDA: move graph and concurrent stream state out of common.cuh (#32001)

am17an committed 1 day ago

Verified 17701e48

ui: use a live-preview markdown editor for the chat form input (#31955)

allozaur committed 1 day ago

Verified ac007054

model : stop renaming the recurrent cache tensors in plamo2 (#31984)

sanmai committed 1 day ago

Verified 940a30ee

common/json-schema : fix allOf with plain types (e.g. integer) generating an empty object grammar (#31834)

chinmaykamjith001 committed 1 day ago

Verified 45c8ea83

Older