ggerganov/llama.cpp

Pull Requests Commits

llama : fix comments for llama_kv_cache API

ggerganov committed 1 year ago

Verified 5418932b

llama : fix handling of "future" tokens when loading sessions

ggerganov committed 1 year ago

Verified 337120cc

llama : temp fix for clearing "future" tokens from the KV cache

ggerganov committed 1 year ago

Verified 0f332a91

Merge branch 'master' into fix-sessions

ggerganov committed 1 year ago

Verified 6a9fe3df

cmake : make CUDA flags more similar to the Makefile (#3420)

cebtenzzre committed 1 year ago

Verified 9476b012

finetune : fix #3404 (#3437)

xaedes committed 1 year ago

Verified a03ce384

metal : set log callback before initializing (#3427)

phronmophobic committed 1 year ago

Verified a8476769

cmake : fix transient definitions in find pkg (#3411)

bandoti committed 1 year ago

Verified 095231df

docker : ignore Git files (#3314)

kevinji committed 1 year ago

Verified ea55295a

infill : add new example + extend server API (#3296)

vvhg1 committed 1 year ago

Verified c97f01c3

ggml-cuda : perform cublas mat mul of quantized types as f16 (#3412)

slaren committed 1 year ago

Verified f5ef5cfb

llama.cpp : add documentation about rope_freq_base and scale values (#3401)

slaren committed 1 year ago

Verified 40e07a60

train : fix KQ_pos allocation (#3392)

ggerganov committed 1 year ago

Verified bc34dd4f

llama : quantize up to 31% faster on Linux and Windows with mmap (#3206)

cebtenzzre committed 1 year ago

Verified 2777a84b

readme : update hot topics + model links (#3399)

BarfingLemurs committed 1 year ago

Verified 0a4a4a09

llama : fix session saving/loading

ggerganov committed 1 year ago

Verified b0670db3

readme : add link to grammars app (#3388)

a10y committed 1 year ago

Verified 569550df

swift : fix build on xcode 15 (#3387)

jhen0409 committed 1 year ago

Verified c71bf2c4

build : enable more non-default compiler warnings (#3200)

cebtenzzre committed 1 year ago

Verified bc39553c

ggml_tensor: update the structure comments. (#3283)

huajsj committed 1 year ago

Verified 0ccfc62a

ggml : release the requested thread pool resource (#3292)

yancaoweidaode committed 1 year ago

Verified 7f1a0fe7

llama.cpp : split llama_context_params into model and context params (#3301)

slaren committed 1 year ago

Verified 16bc66d9

ci : multithreaded builds (#3311)

netrunnereve committed 1 year ago

Verified 0512d666

train : finetune LORA (#2632)

xaedes committed 1 year ago

Verified 0e76a899

gguf : basic type checking in gguf_get_* (#3346)

cebtenzzre committed 1 year ago

Verified 2db94d98

gguf : make token scores and types optional (#3347)

cebtenzzre committed 1 year ago

Verified ecf90b1a

ci : disable freeBSD builds due to lack of VMs (#3381)

ggerganov committed 1 year ago

Verified 2619109a

llama : custom attention mask + parallel decoding + no context swaps (#3228)

ggerganov committed 1 year ago

Verified ec893798

docs : mark code as Bash (#3375)

kevinji committed 1 year ago

Verified 45855b3f

readme : add Mistral AI release 0.1 (#3362)

paschembri committed 1 year ago

Verified 4aea3b84

Older