ngxson/llama.cpp

Pull Requests Commits

vulkan: Fix multi_add invalid descriptor usage (#16899)

jeffbolznv committed 145 days ago

Verified 5d8bb900

vulkan: fuse mul_mat+add and mul_mat_id+add_id (#16868)

jeffbolznv committed 145 days ago

Verified 2e76e013

CUDA: Remove unneded bias/gate dims in fused mmvq (#16858)

ORippler committed 145 days ago

Verified d3dc9dd8

refactor : llama-model.cpp (#16252)

pwilkin committed 145 days ago

Verified bea04522

model : Minimax M2 (#16831)

pwilkin committed 146 days ago

Verified 0de0a015

model : add Granite Hybrid nano types (#16896)

giuseppe committed 146 days ago

Verified e58d5856

CUDA: Volta tensor core support for MMF (#16843)

JohannesGaessler committed 146 days ago

Verified 31c511a9

ggerganov committed 146 days ago

6d39015a

CUDA: add expert reduce kernel (#16857)

am17an committed 146 days ago

Verified 4146d6a1

batch : fix consistency checks for the input positions (#16890)

ggerganov committed 146 days ago

Verified 8da3c0e2

server : don't print user inputs to console (#16871)

ggerganov committed 146 days ago

Verified c22473b5

server : fix typos in server.cpp comments [no ci] (#16883)

danbev committed 146 days ago

Verified 0f715b4e

vulkan: disable spirv-opt for rope shaders (#16872)

jeffbolznv committed 146 days ago

Verified d2d931f1

vulkan: Fix crash when FP16 mul_mat accumulation is not supported (#16796)

rillomas committed 146 days ago

Verified 2976b037

vulkan: fix shmem overrun in mmq id shader (#16873)

0cc4m committed 146 days ago

Verified d2a2673d

ggml-hexagon: respect input size when getting/setting tensor data (#16836)

l3utterfly committed 146 days ago

Verified 13002a08

ci : enable free-disk-space on cuda docker build (#16877)

CISC committed 146 days ago

Verified 6eb208d1

opencl: fix boundary handling for mul_mm (#16875)

lhez committed 146 days ago

Verified 9984cbb6

convert : update transformers requirements (#16866)

RodriMora committed 146 days ago

Verified ce18efea

server : bump request URI max length to 32768 (#16862)

chansikpark committed 147 days ago

Verified 16724b5b

server : remove n_past (#16818)

ggerganov committed 147 days ago

Verified b52edd25

cpu: introduce chunking for repack matmuls and enable matmul-id chunking on ARM64 (#16833)

max-krasnyansky committed 147 days ago

Verified 517b7170

common: fix typo in cli help text (#16864)

sbera77 committed 147 days ago

Verified 835e918d

model: add support for qwen3vl series (#16780)

JJJYmmm committed 147 days ago

Verified d261223d

cpu: introduce chunking for flash attention (#16829)

max-krasnyansky committed 147 days ago

Verified dcca0d3a

model: Add support for CogVLM model (#15002)

Tianyue-Zhao committed 147 days ago

Verified bacddc04

cuda : fix argsort with 64k+ rows (#16849)

CISC committed 147 days ago

Verified 229bf686

llama : use std::abs instead of abs (#16853)

kaetemi committed 147 days ago

Verified d7395115

vulkan: Handle argsort with a large number of rows (#16851)

jeffbolznv committed 147 days ago

Verified 052df28b

Hide latency of bias and gate-loading (#16847)

ORippler committed 147 days ago

Verified 8b11deea

Newer Older