ngxson/llama.cpp

Pull Requests Commits

ci : fine-grant permission

ngxson committed 1 year ago

5422f000

Update README.md (#9591)

32bitmicro committed 1 year ago

Verified 3f1ae2e3

ggerganov committed 1 year ago

Verified f1b8c427

test: fix OPT_STEP_ADAMW for test-backend-ops (ggml/974)

JohannesGaessler committed 1 year ago

Verified e98c1c18

vulkan : mul_mat: fix UB with small warps (ggml/952)

smeso committed 1 year ago

Verified cb000205

ggml : fix ggml_cast (ggml/973)

iboB committed 1 year ago

Verified 6c532248

ggml: fix gradient allocation logic (ggml/966)

JohannesGaessler committed 1 year ago

Verified 7254cdf7

metal : reduce command encoding overhead (#9698)

ggerganov committed 1 year ago

Verified cad341d8

llama : print correct model type for Llama 3.2 1B and 3B

ggerganov committed 1 year ago

Verified a90484c6

convert : refactor rope_freqs generation (#9396)

compilade committed 1 year ago

Verified 1927378b

Fix Docker ROCM builds, use AMDGPU_TARGETS instead of GPU_TARGETS (#9641)

serhii-nakon committed 1 year ago

Verified 6f1d9d71

ci : reduce severity of unused Pyright ignore comments (#9697)

compilade committed 1 year ago

Verified 511636df

py : update transfomers version (#9694)

Vaibhavs10 committed 1 year ago

Verified 08a43d05

flake.lock: Update (#9680)

ggerganov committed 1 year ago

Verified ace4f4be

console : utf-8 fix for windows stdin (#9690)

hasaranga committed 1 year ago

Verified 8277a817

ggml : define missing HWCAP flags (#9684)

ggerganov committed 1 year ago

Verified c919d5db

ggerganov committed 1 year ago

Verified d0b1d663

CUDA: remove bad assert (ggml/972)

JohannesGaessler committed 1 year ago

Verified aaa40999

vulkan : multithread pipeline creation (ggml/963)

jeffbolznv committed 1 year ago

Verified 641002fb

vulkan : fix build for GGML_VULKAN_RUN_TESTS, add TFLOPS to log (ggml/961)

jeffbolznv committed 1 year ago

Verified 0de8b203

vulkan : argsort barriers must be under uniform control flow (ggml/951)

smeso committed 1 year ago

Verified 544f409b

ggml : fix GGML_MAX_N_THREADS + improve formatting (ggml/969)

ggerganov committed 1 year ago

Verified 6084bfb2

common : ensure llama_batch size does not exceed max size (#9668)

matiaslin committed 1 year ago

Verified faac0bae

py : add model class for Chameleon conversion (#9683)

nopperl committed 1 year ago

Verified f99d3f83

contrib : add Resources section (#9675)

ggerganov committed 1 year ago

Verified 589b48d4

llama : add reranking support (#9510)

ggerganov committed 1 year ago

Verified f4d2b884

test-backend-ops : use flops for some performance tests (#9657)

slaren committed 1 year ago

Verified 1b2f992c

llama : add comment about thread-safety [no ci] (#9449)

ggerganov committed 1 year ago

Verified 73984270

vocab : refactor tokenizer to reduce init overhead (#9449)

kylo5aby committed 1 year ago

Verified 6102037b

llama : add support for Chameleon (#8543)

nopperl committed 1 year ago

Verified 9a913110

Older