ggml-org/llama.cpp

Pull Requests Commits

ggml : remove public ggml_backend_buffer_copy_tensor()

ggerganov committed 8 days ago

Verified d72b42fa

cmake: WASI doesn't support spawning (#30751)

MendyBerger committed 8 days ago

Verified e86f5fd1

llama : build the k-pool sparse selection independently of the kq_mask shape (#31243)

am17an committed 8 days ago

Verified 27b7a8a0

rpc : catch std::length_error in recv_msg (#31204)

sohaibharraoui committed 8 days ago

Verified 7ecb57f7

tests : speed up tensor data generation and model saving (#31145)

ggerganov committed 8 days ago

Verified 8542e943

vulkan: fix CI llvmpipe by allowing dim 3 broadcast in MUL_MAT (#31239)

ServeurpersoCom committed 8 days ago

Verified 922f53ed

mtmd-cli: Reduced scope of batch pointer (#31155) (#31165)

SamMalayek committed 8 days ago

Verified 0c8c59ef

ui: fix system message row stretching to full prompt height in WebKit (#31246)

allozaur committed 8 days ago

Verified 3ab37aed

server: add router device memory margin parameter for dynamic unloading (#21231)

0cc4m committed 8 days ago

Verified 85cc432e

common : --model or -hf is required (#31242)

julien-c committed 8 days ago

Verified 74a1fd3d

model: Add MiniCPM-OCR-Jev (#30980)

tc-mb committed 8 days ago

Verified 7eed6b76

metal : add GGML_OP_TOP_K_MASK (#31230)

ggerganov committed 8 days ago

Verified 27637851

glm5-next : QSA without kq_mask (+ DSA) (#31009)

sanmai committed 8 days ago

Verified d78f40d9

ci : use ccache depend mode for ROCm release builds (#31231)

sreeram-11 committed 8 days ago

Verified fae9ac56

ci : force python 3.12 on ubuntu 26.04 (#31222)

CISC committed 8 days ago

Verified a8276b70

ggml : relax scale to non-contiguous views (#31146)

ggerganov committed 8 days ago

Verified de5122d8

tests : fix multi-seq split replay and partial-state cases for trained models (#31216)

ggerganov committed 8 days ago

Verified 6666a733

llama : remove obsolete n_pad from memory instances (#31200)

ggerganov committed 8 days ago

Verified 7a2e5aba

ui: list org-less models first in the model selector (#31215)

allozaur committed 8 days ago

Verified 0d5ca600

opencl: refine mul mat id -1 index (#31170)

lhez committed 8 days ago

Verified d930fe3c

opencl: initial support for q3_k (#31172)

lhez committed 8 days ago

Verified a2b3e121

ui: Highlight reasoning toggle on hover and align it with model selector (#31212)

allozaur committed 8 days ago

Verified 823df7d0

ui: add conversation compaction (#27401)

ServeurpersoCom committed 8 days ago

Verified 1a97e387

ui: Open model details in a drawer without leaving the page (#31176)

allozaur committed 8 days ago

Verified bc262f33

ui: use shared confirmation dialog for conversation deletion (#31187)

allozaur committed 8 days ago

Verified f5892ca5

hexagon: add GGML_HEXAGON_VTCM_SIZE (#31221)

ebateni committed 8 days ago

Verified 7aca0d2e

llama : enable moe amax rescaling for laguna (#31147)

begcdn committed 8 days ago

Verified 961c376a

ci : add windows behavior check (#31191)

CISC committed 8 days ago

Verified 4244ad7a

opencl: write KV-cache views in set_tensor so a state restore is not dropped (#30471)

wanghqc committed 9 days ago

Verified 63b4b303

hexagon: MUL_MAT_ID row skipping optimizations and improvements (#31169)

max-krasnyansky committed 9 days ago

Verified caa8191d

Older