Pull Requests ggerganov/llama.cpp

[MODEL] support qwen3.5 series

#19468 opened 2026-02-09 19:02 by JJJYmmm

test: fix IMROPE perf test case testing

#19465 opened 2026-02-09 15:38 by ngxson

tts : fix typos in README.md [no ci] examples

#19463 opened 2026-02-09 15:02 by danbev

llama : use n_embd_features for input embedding dimension if specified

#19462 opened 2026-02-09 14:58 by danbev

Add a workaround for compilation with ROCWMMA_FATTN and gfx9 Nvidia GPU ggml

#19461 opened 2026-02-09 14:52 by superm1

model: support GLM MoE DSA arch model python

#19460 opened 2026-02-09 14:41 by ngxson

python: Use NumPy 2.0+ examples python server

#19455 opened 2026-02-09 13:01 by akx

ggml: use noexcept overload for is_regular_file in backend registration ggml

#19452 opened 2026-02-09 10:57 by k4ss4n

Added HVX support for SQR,SQRT,DIV,SUM_ROWS,ARGSORT ggml

#19448 opened 2026-02-09 10:35 by YardenTal44

Add special case for Qwen3VLMoe python

#19445 opened 2026-02-09 09:15 by pwilkin

mtmd: (WIP) qwen3 audio support examples python

#19441 opened 2026-02-08 23:26 by ngxson

Use Vulkan SDK Constants in ggml-vulkan.cpp Vulkan ggml

#19440 opened 2026-02-08 21:56 by inforithmics

tools: add quant-bench for profiling raw kernel performance examples

#19434 opened 2026-02-08 17:27 by chethanreddy1

Add a build target to generate ROCm artifacts using ROCm 7.2 devops

#19433 opened 2026-02-08 15:32 by superm1

cuda : extend GGML_OP_PAD to work with non-cont src0 testing Nvidia GPU ggml

#19429 opened 2026-02-08 11:25 by ggerganov

Updated documentation

#19428 opened 2026-02-08 09:44 by MonkeybreadSoftware

metal : fix ACC op ggml Apple Metal

#19427 opened 2026-02-08 09:09 by ggerganov

fix vulkan ggml_acc only works in 3d but not 4d testing Vulkan ggml

#19426 opened 2026-02-08 08:23 by ymcki

ggml-cpu: FA add GEMM microkernel ggml

#19422 opened 2026-02-07 19:02 by am17an

Update ROCm docker container to 7.2 devops

#19418 opened 2026-02-07 14:35 by superm1

fix: correct typos 'occured' and 'occurences'

#19414 opened 2026-02-07 11:44 by thecaptain789

Add compiler flags for UWP

#19413 opened 2026-02-07 11:17 by Iemand005

sampling : blue noise rng examples server

#19409 opened 2026-02-07 07:42 by kaetemi

hexagon: further optimization and tuning of matmul and dot kernels ggml

#19407 opened 2026-02-07 01:35 by max-krasnyansky

hexagon: Add ARGSORT, DIV, SQR, SQRT, SUM_ROWS, GEGLU ggml

#19406 opened 2026-02-07 01:11 by max-krasnyansky

opencl: refactor expm1 and softplus ggml OpenCL

#19404 opened 2026-02-06 23:57 by shaofeiqi

ggml-cpu: optimize ggml_vec_dot_bf16 for s390x ggml

#19399 opened 2026-02-06 17:59 by taronaeo

ggml: backend-agnostic tensor parallelism Nvidia GPU Vulkan examples ggml SYCL Apple Metal Ascend NPU OpenCL IBM zDNN

#19378 opened 2026-02-05 22:30 by JohannesGaessler

models : optimizing qwen3next graph model

#19375 opened 2026-02-05 20:57 by ggerganov

WebUI hide models in router mode examples server

#19374 opened 2026-02-05 20:39 by crsawyer