ggml-org/llama.cpp

Pull Requests Commits

cpu: add bf16 support for scale, clamp, leaky_relu, swiglu_oai, swiglu_clamp, tri and diag

0cc4m committed 1 day ago

211385ca

vulkan: expand bf16 support for unary and binary ops and rms_norm

0cc4m committed 1 day ago

8d191e35

ggml : add BF16 support to RMS_NORM (CPU, CUDA) (#32135)

anujj committed 1 day ago

Verified 31953bcf

vulkan: add TOP_K_MASK (#31544)

developer0hye committed 1 day ago

Verified ae93877c

spec : Warn when --spec-draft-p-min is used with probabilistic draft sampling (#31294)

praneshgo committed 1 day ago

Verified 7874988b

ggml-vulkan: add optional device keepalive (#29267)

ac-mmi committed 1 day ago

Verified 8c2c4c39

grammar : optimize matching with left factoring (#29557)

o7si committed 1 day ago

Verified 108f73ad

llama: move moe-cache code out of llama-graph (#32157)

am17an committed 1 day ago

Verified 951d465f

CUDA: split the PAD_REFLECT_1D grid to stay within the HIP grid limit (#32179)

ServeurpersoCom committed 1 day ago

Verified 2d8a3250

examples : remove explict version from test-cmake (#32191)

danbev committed 1 day ago

Verified b1b770d9

llama.cpp : bump version to 1.0.0 (#32190)

ggerganov committed 1 day ago

Verified 4cda6227

ggerganov committed 1 day ago

68598133

ggml : bump version to 1.0.0 (ggml/1690)

ggerganov committed 1 day ago

3bfde28f

examples : clean-up old/obsolete stuff (#32184)

ggerganov committed 1 day ago

Verified 28d37fe2

ci: fix HIP quality check by ignoring the FP8 vec FA kernel spill (#32183)

ServeurpersoCom committed 1 day ago

Verified bd2767d0

ci : set windows-cuda ccache size to 1G (#32185)

CISC committed 1 day ago

Verified 82503b42

chat : accept </ifm|think> as K2 Horizon closer at medium/low effort (#30335)

awss1i committed 1 day ago

Verified 5676dfbe

server, fix CI: read a small temp file in the read_file tool test [no release] (#32168)

ServeurpersoCom committed 1 day ago

Verified 0a66ac71

server: allow private media URLs in the vision tests [no ci] (#32167)

ServeurpersoCom committed 1 day ago

Verified 892b40b9

server: skip LAN address tests when the host cannot reach it [no ci] (#32166)

ServeurpersoCom committed 1 day ago

Verified e9701014

fix(hexagon): make host build work with clang-cl on Windows ARM64 (#32011)

SaidSultani committed 1 day ago

Verified 92d6c8a5

llama : prune (#30565)

ggerganov committed 2 days ago

Verified 8ca5a800

ggml : prune (#29801)

ggerganov committed 2 days ago

Verified 29dbbad2

hexagon: add Q5_0 and Q5_1 support (#32094)

aparmp-quic committed 2 days ago

Verified 2a4daf0e

server : cancel MCP tool calls on client disconnect or timeout (#32025)

hmblair committed 2 days ago

Verified d8c8cc4d

qwen3.5: merge tensors on convert also (#32102)

ORippler committed 2 days ago

Verified 3ba1ce0e

Title: CUDA: few-row MMQ for BS 2-24 (#31533)

anujj committed 2 days ago

Verified 0fb8f157

ci : build ui on non-slim runner (#32138)

CISC committed 2 days ago

Verified 827466b9

server: fix SSRF for CORS proxy and media URLs (#32115)

ngxson committed 2 days ago

Verified 8954f194

server: explicit opt-in tools via non-loopback addr, check cors host port (#32113)

ngxson committed 2 days ago

Verified a1a68eb7

Older