ggml-org/llama.cpp

Pull Requests Commits

convert-hf : adapt ArcticModel to use yield too

compilade committed 1 year ago

11f78c6a

Merge branch 'master' into compilade/lazier-moe-convert-hf

compilade committed 1 year ago

96a299ff

convert-hf : fix flake8 indentation lint

compilade committed 1 year ago

d703fa9f

train : change default FA argument (#7528)

ggerganov committed 1 year ago

Verified 9588f196

labeler: added Apple Metal detector (+Kompute) (#7529)

mofosyne committed 1 year ago

Verified 3cbd23ed

main : don't print special tokens with --grammar (#6923)

jart committed 1 year ago

Verified 00c63907

ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)

msy-kato committed 1 year ago

Verified faa0e697

android : module (#7502)

eltonkola committed 1 year ago

Verified 9791f402

fix missing slash in `fs_get_cache_directory()` (#7503)

ngxson committed 1 year ago

Verified 902184dd

Make tokenize CLI tool have nicer command line arguments. (#6188)

Noeda committed 1 year ago

Verified 57684331

gguf-py : fix and simplify quantized shape round-trip (#7483)

compilade committed 1 year ago

Verified b83bab15

flake.lock: Update (#7232)

ggerganov committed 1 year ago

Verified d041d2ce

docker.yml: disable light-intel and server-intel test (#7515)

mofosyne committed 1 year ago

Verified 27891f6d

Add support for ArcticForCausalLM (#7020)

fairydreaming committed 1 year ago

Verified fbca2f27

add build shared lib in win release package (#7438)

arthw committed 1 year ago

Verified 0df0aa8e

readme : remove trailing space (#7469)

ggerganov committed 1 year ago

Verified 74f33adf

ggml : silence UB sanitizer error during iq2_xxs quantization (#0)

ggerganov committed 1 year ago

Verified 1debe727

Fix phi3 chat template confusion with zephyr (#7449)

tristandruyen committed 1 year ago

Verified 007489e8

readme : add Bunny in supported models [no ci] (#7469)

criminact committed 1 year ago

Verified 8b94e799

llama : add getters for n_threads/n_threads_batch (#7464)

danbev committed 1 year ago

Verified 3015851c

ci : use Pythia models instead of OpenLlama (#7470)

ggerganov committed 1 year ago

Verified 55ac3b7a

readme : add GPT-NeoX + Pythia to the list of supported models (#7491)

felladrin committed 1 year ago

Verified dacfcebd

Add missing inference support for GPTNeoXForCausalLM (Pythia and GPT-NeoX base models) (#7461)

fairydreaming committed 1 year ago

Verified 9b82476e

llama : rename n_ctx -> cache.size, less confusing (#0)

ggerganov committed 1 year ago

Verified a61a94e5

labeler.yml: add embedding label detector [no ci] (#7482)

mofosyne committed 1 year ago

Verified 152da28a

ggml : remove ggml_flash_attn and ggml_flash_ff (#7463)

ggerganov committed 1 year ago

Verified d48c88cb

ggml : drop support for QK_K=64 (#7473)

ggerganov committed 1 year ago

Verified e84b71c2

Update vulkan rope implementation to support frequency factors (#7475)

0cc4m committed 1 year ago

Verified 1b1e27cb

main : minor (#7462)

ggerganov committed 1 year ago

Verified fbf777d2

CUDA: fix FA out-of-bounds reads (#7479)

JohannesGaessler committed 1 year ago

Verified cd93a28c

Older