ochafik/llama.cpp

Pull Requests Commits

chat: fix jinja parser for TranslateGemma (#30096)

nbud committed 14 days ago

Verified bd4eeaa0

chat : name tool and argument parser rules by index (#30088)

Frost-54 committed 14 days ago

Verified 5de73343

model : add LiquidAI/d1-3B decision model (#30110)

tdakhran committed 14 days ago

Verified 88dcc460

cuda: FWHT kernels for block widths above 512 (#29100)

bri-prism committed 14 days ago

Verified b86d2f07

mtmd: add cohere2 vision support (#30062)

Terrencezzj committed 14 days ago

Verified 50a6c5cf

llama : add a GPU cache for MoE experts kept in host memory (#29887)

am17an committed 14 days ago

Verified d6cf9acb

cuda: update uncoalesced memory reads in pool2d (#29425)

shenron0101 committed 14 days ago

Verified 42c787e8

server : accumulate generated text and tokens as parse input (#29876)

aldehir committed 14 days ago

Verified 18b5f8b1

llama: share the nextn tensor flags between models (#30097)

ServeurpersoCom committed 15 days ago

Verified 448147d4

metal : few-row MMA mat-mul for the remaining src0 types (#30065)

pratiknarola-t committed 15 days ago

Verified 98819068

metal : fix MUL_MAT+ADD fusion when the residual is itself a MUL_MAT (#30100)

nyo16 committed 15 days ago

Verified 7e8324f5

feat: add GLM5Next MTP, optimize (#29928)

pwilkin committed 15 days ago

Verified b9acf138

qwen3tts : guard speaker_encoder_config patch for CustomVoice variant (#29179)

SIDDARTHAREDDY8 committed 15 days ago

Verified 48499d2e

sampling : use greedy selection for eligible temperature-zero chains (#29797)

hthadicherla committed 15 days ago

Verified d0b490f2

vocab : add plamo fim tokens (#30090)

CISC committed 15 days ago

Verified 42b021b4

convert : fix token configuration for PLaMo-3 (#29843)

tokinasin committed 15 days ago

Verified 7481354a

musa: use the tile lightning indexer kernel (#30080)

yeahdongcn committed 15 days ago

Verified ad215653

vendor : update cpp-httplib to 0.60.0 (#30081)

cabelo committed 15 days ago

Verified b7dafa01

imatrix : include clocale for std::setlocale (#30079)

dazzywi committed 15 days ago

Verified 26908739

ggml-webgpu: fix flash_attn supports_op check for overlapping KV (#28205)

yomaytk committed 15 days ago

Verified 36a73916

tests : retain the anchor when testing recurrent rollback (#29923)

eapache committed 15 days ago

Verified fa3c2fab

[SYCL] fix the issue in mixed different model GPUs in FA (#29071)

arthw committed 15 days ago

Verified 005a1e12

sycl: accelerate GLM MLA prefill with MKL flash attention (#29171)

anantshri committed 15 days ago

Verified d2a79e60

sycl : fattn_kv_buffers cleanup (#27689)

Pratyush-gg committed 15 days ago

Verified 5e5b628e

vulkan: fix amd iGPU slow checkpoint read (#30049)

0cc4m committed 15 days ago

Verified 4d756bc7

sycl: add IQ3_S multi-column MMVQ (#29500)

clemenswasser committed 15 days ago

Verified 78651c41

ggml-webgpu: no dawn native features on wasi (#27069)

MendyBerger committed 15 days ago

Verified f498f864

hexagon: CPY/CONCAT/CONT/DUP overhaul to use DMA/HVX for all cases (#30067)

max-krasnyansky committed 15 days ago

Verified c479922a

cuda : add BF16 support for XIELU (#29955)

Qiao12-pixel committed 15 days ago

Verified 5ad1c5da

ggml-cuda: use per-thread stream for buffer-init padding memset (#28782)

harkgill-amd committed 15 days ago

Verified 51ce9c11

Older