ngxson/llama.cpp

Pull Requests Commits

helper: use mtmd_image_tokens_get_n_pos

ngxson committed 250 days ago

076e3b93

add warning about breaking changes

ngxson committed 250 days ago

049ae242

add const to various places

ngxson committed 250 days ago

e9f7ff92

Merge branch 'master' into xsn/mtmd_c_api

ngxson committed 250 days ago

06cb5952

vulkan: Additional type support for unary, binary, and copy (#13266)

jeffbolznv committed 250 days ago

Verified 8ae5ebcf

imatrix: fix oob writes if src1 is not contiguous (#13286)

JohannesGaessler committed 251 days ago

Verified 3e959f09

clip : revert the change of BOI/EOI token for GLM-edge (⚠️ breaking change) (#13259)

ngxson committed 251 days ago

Verified 36667c8e

llama : Llama-3_1-Nemotron-Ultra-253B-v1 support (#12843)

ymcki committed 251 days ago

Verified 3bf785f3

ngxson committed 252 days ago

4d842eb9

mtmd_input_chunk_copy

ngxson committed 252 days ago

6bc7a30a

explicitly create mtmd::input_chunks

ngxson committed 252 days ago

a0fb7016

fix merging issue

ngxson committed 252 days ago

863db315

Merge branch 'master' into xsn/mtmd_c_api

ngxson committed 252 days ago

a230804b

llama : move end-user examples to tools directory (#13249)

slaren committed 252 days ago

Verified 1d36b367

sync : ggml (#13268)

ggerganov committed 252 days ago

Verified b3444392

context : fix reorder logic (#13267)

ggerganov committed 252 days ago

Verified a75cb30d

ggml : Enable MMA for BF16 in llamafile_sgemm (#13148)

shalinib-ibm committed 252 days ago

Verified 3f3769ba

llama-model : support Qwen2 embedding models and pooling_mode_lasttoken (#13245)

cebtenzzre committed 252 days ago

Verified 2f567611

convert : use correct context length for nomic-embed-text-v2 (#13216)

cebtenzzre committed 252 days ago

Verified 7d212348

convert : converting mmproj for Qwen2/2.5VL from convert_hf_to_gguf (#13209)

ngxson committed 252 days ago

Verified 074e42ab

kv-cache : separate recurrent vs non-recurrent impl (#12799)

ggerganov committed 252 days ago

Verified c642bc01

llama : orion rope type is neox (#13261)

CISC committed 252 days ago

Verified cb06a3c3

llama : plamo rope type is neox (#13260)

CISC committed 252 days ago

Verified 626083fa

llama-chat : reset glmedge chat template (#13253)

piDack committed 252 days ago

Verified 2af68801

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

ahmedshakill committed 252 days ago

Verified e84773ab

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

ggerganov committed 252 days ago

Verified fab647e8

convert : explicitly disable trust_remote_code for AutoConfig (#13246)

ngxson committed 252 days ago

Verified dcf88600

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

bandoti committed 253 days ago

Verified d24d5928

rpc : avoid uninitialized memory in serialize_tensor (#13210)

justinsb committed 253 days ago

Verified 8efbdadc

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

jessegross committed 253 days ago

Verified f057808f

Older