ggerganov/llama.cpp

Pull Requests Commits

fix and opt kernel launch

rmatif committed 112 days ago

f68669d5

fix trailing whitespace

rmatif committed 116 days ago

18310bf2

add mul_mat_f16_f32_image kernel

rmatif committed 116 days ago

418606e8

vocab : add midm-2.0 model pre-tokenizer (#14626)

Bing-su committed 116 days ago

Verified 576c82ed

model : Granite Four (#13550)

gabe-l-hart committed 117 days ago

Verified 0aedae00

opencl: add tiled mul_mat_f16_f32 (#14535)

rmatif committed 117 days ago

Verified 6bdda139

opencl: add `set_rows` for `f16` and `f32` (#14547)

lhez committed 117 days ago

Verified 0b885577

Smoldocling support (#14597)

ryan-mangeno committed 117 days ago

Verified 4bb625b7

Docs: script to auto-generate ggml operations docs (#14598)

am17an committed 117 days ago

Verified 11ee0fea

cmake : do not search for curl libraries by ourselves (#14613)

EZForever committed 117 days ago

Verified a4575513

SYCL: Initial set_rows kernel implementation (#14562)

qnixsynapse committed 117 days ago

Verified 704bb7a7

llama : minor coding style fix for smollm3 (#14605)

ngxson committed 118 days ago

Verified 435a6d10

cmake : bump llguidance version to v1.0.1 (#14609)

EZForever committed 118 days ago

Verified f9a867f5

cmake : llguidance build parser library only (#14608)

EZForever committed 118 days ago

Verified ac44eb6c

cuda : support Falcon-H1 state size for SSM_SCAN (#14602)

compilade committed 118 days ago

Verified a57d1bcb

llama : remove llm_graph_input_one (#14603)

ngxson committed 118 days ago

Verified cb9178f8

llama : support Jamba hybrid Transformer-Mamba models (#7531)

compilade committed 118 days ago

Verified 4a5686da

ggml : add ggml_scale_bias (#14417)

ngxson committed 118 days ago

Verified 98bab638

ggml : prevent integer overflow in gguf tensor size calculation (#14595)

Yuuoniy committed 118 days ago

Verified 26a48ad6

model : add skt/A.X-4.0 model vocabulary (#14589)

Bing-su committed 118 days ago

Verified ffd59e7d

llama : remove unintended whitespace (#14592)

CISC committed 118 days ago

Verified 10555459

model : add support for Falcon-H1 family (#14534)

ibrahimkhadraoui committed 118 days ago

Verified 04655063

convert : fix smollm3 jinja template (#14586)

ngxson committed 119 days ago

Verified 20b7bf8a

vulkan: optimize flash attention split_k_reduce (#14554)

jeffbolznv committed 119 days ago

Verified 6efcd659

model : fix hunyuan moe chat template (#14584)

stevenkuang-tencent committed 119 days ago

Verified 699f4392

model : add SmolLM3 (#14581)

ngxson committed 119 days ago

Verified 08382869

memory : fix broken batch splits for recurrent cache (#14575)

compilade committed 119 days ago

Verified bb4f7a9e

vulkan : fix rope with partial rotation and non-cont src (#14582)

jeffbolznv committed 119 days ago

Verified b8eeb874

server: Add ability to mount server at prefix (#14544)

oluwandabira committed 119 days ago

Verified 17a1f0d2

model : add hunyuan moe (#14425)

ngxson committed 119 days ago

Verified 8f22dc0a

Older