ggerganov/llama.cpp

Pull Requests Commits

ggml : fix handling of zero blocks in IQ quants

ggerganov committed 339 days ago

Verified 28f7a4d0

Vulkan Shader Refactor, Memory Debugging Option (#7947)

0cc4m committed 339 days ago

Verified 7c7836d9

Add `cvector-generator` example (#7514)

ngxson committed 340 days ago

Verified 0c7b3595

[SYCL] remove global variables (#7710)

airMeng committed 340 days ago

Verified 7b2f4a7d

ci : fix macos x86 build (#7940)

olexiyb committed 341 days ago

Verified f8ec8877

CUDA: faster q2_K, q3_K MMQ + int8 tensor cores (#7921)

JohannesGaessler committed 341 days ago

Verified 76d66ee0

metal : utilize max shared memory for mul_mat_id (#7935)

ggerganov committed 341 days ago

Verified 66ef1cee

llama-bench : fix RPC indication (#7936)

rgerganov committed 341 days ago

Verified e65bbf60

llama : more checks before assuming FIM tokens (#7644)

CISC committed 341 days ago

Verified 6fcd1331

convert : add Poro-34B-chat tokenizer support (#7713)

ezosa committed 341 days ago

Verified 41b9260f

rpc : fix ggml_backend_rpc_supports_buft() (#7918)

rgerganov committed 342 days ago

Verified 172c8256

readme : Remove outdated instructions from README.md (#7914) [no ci]

Galunid committed 342 days ago

Verified a55eb1bf

move BLAS to a separate backend (#6210)

slaren committed 342 days ago

Verified f578b86b

`build`: rename main → llama-cli, server → llama-server, llava-cli → llama-llava-cli, etc... (#7809)

ochafik committed 342 days ago

Verified 1c641e6a

CUDA: fix broken oob check for FA vec f32 kernel (#7904)

JohannesGaessler committed 343 days ago

Verified 96355290

tests : add non-cont unary tests (#7857)

ggerganov committed 343 days ago

Verified a9cae480

ggml : improve ggml_is_contiguous logic (#7856)

ggerganov committed 343 days ago

Verified bfaa676b

server : restore numeric prompts (#7883)

ggerganov committed 343 days ago

Verified 704a35b1

update intel docker oneapi-basekit to 2024.1.1-devel-ubuntu22.04 (#7894)

airMeng committed 343 days ago

Verified dcf75270

Fix a typo and add Fedora 40 pacakge to install for Vulkan (#7794) [no ci]

metal3d committed 343 days ago

Verified f2b5764b

vulkan: select only one device for single gpu with multiple drivers (#7582)

Adriankhl committed 344 days ago

Verified 73bac2b1

Update Vulkan RoPE implementation (#7818)

0cc4m committed 344 days ago

Verified ef52d1d1

fix broken link in pr template (#7880) [no ci]

deven367 committed 344 days ago

Verified 14f83526

github: move PR template to .github/ root (#7868)

mofosyne committed 344 days ago

Verified 6fe42d07

llama-bench: more compact markdown tables (#7879)

JohannesGaessler committed 344 days ago

Verified 148995e5

tests : check the Python version (#7872)

ggerganov committed 344 days ago

Verified 4bfe50f7

CUDA: int8 tensor cores for MMQ (q4_K, q5_K, q6_K) (#7860)

JohannesGaessler committed 344 days ago

Verified bdcb8f42

fix CUDA CI by using a windows-2019 image (#7861)

slaren committed 344 days ago

Verified c2ce6c47

json: refine constraint for whitespace to avoid runaways yet allow pretty print (#7866)

ochafik committed 344 days ago

Verified b61eb964

`json`: document schema conversion in GBNF readme, align manual grammar examples & converters (#7841)

ochafik committed 344 days ago

Verified 396b18df

Older