ggerganov/llama.cpp

Pull Requests Commits

fix speculative decoding build on windows (#5874)

jquesnelle committed 1 year ago

Verified 29eee404

nix: static build (#5814)

hutli committed 1 year ago

Verified 1d41d6f7

llama : fix embeddings (#5796)

ggerganov committed 1 year ago

Verified 29ae62d2

ggerganov committed 1 year ago

Verified e0843afe

ggml : fix unknown status (#0)

ggerganov committed 1 year ago

a1c6d96e

ggerganov committed 1 year ago

efd8533e

ggml : introduce ggml_status (ggml/750)

Xarbirus committed 1 year ago

9fa26273

cmake : handle cases where git index is not found in .git (#5844)

danemadsen committed 1 year ago

Verified fe52be11

speculative : implement stochastic speculative sampling (#5625)

mscheong01 committed 1 year ago

Verified 6d341ab6

add alias for chat template (#5858)

ngxson committed 1 year ago

Verified 4ffcdce2

ggerganov committed 1 year ago

Verified a0fc6266

add some new ops, fix some operators and add batch operations to certain operators. (ggml/747)

leejet committed 1 year ago

Verified 7d43c585

common : use LLAMA_DEFAULT_SEED (#5855)

dranger003 committed 1 year ago

Verified 82f3e668

main : support special tokens as reverse/anti prompt (#5847)

dranger003 committed 1 year ago

Verified 5a51cc1b

cuda : fix data race in soft max (#5853)

slaren committed 1 year ago

Verified 67be2ce1

readme : add API changes section

ggerganov committed 1 year ago

Verified 231ae28f

llama : allow for user specified embedding pooling type (#5849)

iamlemec committed 1 year ago

Verified 475df1d6

gguf-dump : support i-quants (#5841)

Nindaleth committed 1 year ago

Verified 87c2e8b2

llama : fix llama_copy_state_data with fragmented KV cache (#5840)

compilade committed 1 year ago

Verified de9692a7

ci : schedule slow server tests only on Release or on demand (#5839)

phymbert committed 1 year ago

Verified e6029348

server : init http requests thread pool with --parallel if set (#5836)

phymbert committed 1 year ago

Verified 8ef969af

flake.lock: Update (#5842)

ggerganov committed 1 year ago

Verified fa974646

server: tests: passkey challenge / self-extend with context shift demo (#5832)

phymbert committed 1 year ago

Verified 97311342

llama : add abort_callback to interrupt computation (#5409)

Xarbirus committed 1 year ago

Verified 4a6e2d61

ggml : fix IQ3_S AVX implementation (#5834)

ggerganov committed 1 year ago

Verified 494c8703

convert : automatically fall back to HfVocab if tokenizer.model doesn't exist (#5821)

cebtenzzre committed 1 year ago

Verified 4d4d2366

convert-hf : make model class definitions self-contained (#5825)

cebtenzzre committed 1 year ago

Verified c7a0ad8e

ggml : IQ3_S improvements (#5829)

ikawrakow committed 1 year ago

Verified bbde6eb2

scripts : add pod-llama.sh

ggerganov committed 1 year ago

Verified ef2cd694

llama : refactor internal quantization functions (#5830)

ngxson committed 1 year ago

Verified 6c32d8c7

Newer Older