ngxson/llama.cpp

Pull Requests Commits

move V_ENC_EMBD_PATCH handling to Qwen2VLVisionModel

ngxson committed 350 days ago

c030984b

Merge branch 'master' into xsn/convert_gguf_qwen2vl

ngxson committed 350 days ago

62b7b1e7

mtmd-cli : fix out_of_range when input image path is empty (#13244)

ahmedshakill committed 350 days ago

Verified e84773ab

server : add cache reuse card link to help (#13230)

ggerganov committed 351 days ago

Verified fab647e8

convert : explicitly disable trust_remote_code for AutoConfig (#13246)

ngxson committed 351 days ago

Verified dcf88600

ci: fix cross-compile sync issues (#12804)

mtmcp committed 351 days ago

Verified d24d5928

rpc : avoid uninitialized memory in serialize_tensor (#13210)

justinsb committed 351 days ago

Verified 8efbdadc

ggml: Don't assert fail when tensor data changes (#13222)

jessegross committed 351 days ago

Verified f057808f

build : fix build info on windows (#13239)

slaren committed 351 days ago

Verified d7a14c42

clip : (minicpmv) Re-enable upscaling of images smaller than the CLIP image size (#13237)

lcarrere committed 351 days ago

Verified b6e4ff69

llama-chat : update GLM4 chat template (#13238)

matteoserva committed 351 days ago

Verified e0f572c8

vulkan: Add bfloat16 support (#12554)

jeffbolznv committed 351 days ago

Verified 79f26e9e

vulkan: Handle src1 batch dimension in non-contiguous mat-vec-mul shader (#13191)

jeffbolznv committed 351 days ago

Verified fc727bcd

test: non-cont. b in test-backend-ops -o MUL_MAT (#13187)

JohannesGaessler committed 351 days ago

Verified b0ecbd43

ggerganov committed 351 days ago

b1dd4d08

whisper : add check that target name exists (whisper/3103)

danbev committed 351 days ago

99881f77

ggml : suppress Windows compiler warnings (whisper/3075)

danbev committed 351 days ago

b5769d92

mtmd : add **vision** support for Mistral Small 3.1 (#13231)

ngxson committed 351 days ago

Verified 8936784f

arg : remove CURLINFO_EFFECTIVE_METHOD (#13228)

ngxson committed 351 days ago

Verified 13c9a331

llama-model : fix the reported size class for nomic-embed-text-v2-moe (#13223)

cebtenzzre committed 351 days ago

Verified a70183eb

ggerganov committed 352 days ago

8d33d740

ggml : fix ggml_gallocr_ptr type (ggml/1205)

slaren committed 352 days ago

4254bb49

cuda : fix unused variable compile warning (whisper/0)

ggerganov committed 352 days ago

99985401

CUDA: batched+noncont MMQ, refactor bs>1 MoE code (#13199)

JohannesGaessler committed 352 days ago

Verified e1e8e099

update llava/readme

ngxson committed 352 days ago

ef0bc7aa

rm qwen surgery script

ngxson committed 352 days ago

6e31ddcd

clarify about 32B

ngxson committed 352 days ago

13e4cccb

Revert "fix 32B model"

ngxson committed 352 days ago

d96ef53a

arg : -hf do not fail if url mismatch (#13219)

ngxson committed 352 days ago

Verified 6f67cf1f

fix typo: `n_ctx_pre_seq` -> `n_ctx_per_seq` (#13221)

ddh0 committed 352 days ago

Verified 16a457fa

Older