ngxson/llama.cpp

Pull Requests Commits

add MTMD_API to fix build

ngxson committed 9 days ago

388b674d

rm legacy clip_debug_encode()

ngxson committed 10 days ago

96799a48

fix windows compile error

ngxson committed 10 days ago

768c6f69

fix compile error

ngxson committed 10 days ago

0765d434

ngxson committed 10 days ago

dc7b1831

ngxson committed 10 days ago

12ace406

Merge branch 'master' into xsn/mtmd_debug

ngxson committed 10 days ago

2f5977f2

mtmd: add llama-mtmd-debug binary

ngxson committed 10 days ago

ad35ea9e

mtmd : rename mtmd_get_audio_bitrate to mtmd_get_audio_sample_rate (#20105)

danbev committed 10 days ago

Verified 8f974d23

general: CONTRIBUTING.md - guidelines for quantization schemes (#19762)

pwilkin committed 10 days ago

Verified 2948e604

metal : fix l2 norm scale (#20493)

ggerganov committed 10 days ago

Verified 73c9eb8c

convert : fix/suppress pyright errors (#20442)

danbev committed 11 days ago

Verified 983df142

llama : disable graph reuse with pipeline parallelism (#20463)

ggerganov committed 11 days ago

Verified 57819b8d

vendor : update cpp-httplib to 0.37.1 (#20390)

cabelo committed 11 days ago

Verified 557fe2d9

tests : use `reasoning` instead of `reasoning_budget` in server tests (#20432)

pwilkin committed 11 days ago

Verified 0e810413

test-backend-ops: allow loading tests from file and parsing model operators into file (#19896)

0cc4m committed 11 days ago

Verified 128142fe

common : update completion executables list [no ci] (#19934)

danbev committed 11 days ago

Verified 6de1bc63

grammar: Fix grammar root symbol check (#19761)

AsbjornOlling committed 11 days ago

Verified 0a10c34d

vulkan: add GATED_DELTA_NET op support (#20334)

ProgenyAlpha committed 11 days ago

Verified deee2386

convert : better mtp check and fix return [no ci] (#20419)

CISC committed 11 days ago

Verified c3e3f9e5

vulkan: fix SSM_CONV PP scaling with large ubatch sizes (#20379)

ProgenyAlpha committed 11 days ago

Verified 40c550d4

New conversations now auto-select the first loaded model (#20403)

ServeurpersoCom committed 11 days ago

Verified de190154

ggml-virtgpu: Fix some build commands (#20341)

yomaytk committed 11 days ago

Verified 05039967

metal : avoid divisions in bin kernel (#20426)

ggerganov committed 11 days ago

Verified e4cff095

ci: Setup self-hosted CI for Intel Linux Vulkan backend (#20154)

rillomas committed 11 days ago

Verified 4cc6eb15

vulkan: fix l2_norm epsilon handling (#20350)

jeffbolznv committed 11 days ago

Verified 246ffc4b

vulkan: fix OOB check in flash_attn_mask_opt (#20296)

jeffbolznv committed 11 days ago

Verified aa429cf5

vulkan: Fix ErrorOutOfHostMemory on Intel GPU when loading large models with --no-mmap (#20059)

rillomas committed 12 days ago

Verified 5866e3bb

opencl: use larger workgroup size for get_rows (#20316)

lhez committed 12 days ago

Verified 0516e04b

opencl: add cumsum op (#18981)

shaofeiqi committed 12 days ago

Verified 3d9ab225

Older