ggerganov/llama.cpp

Pull Requests Commits

server : add comment about changing slot_state to bool

ggerganov committed 1 year ago

Verified c0f4d548

server : fix slot reuse

ggerganov committed 1 year ago

Verified 83e14901

server : apply fix from #3722

ggerganov committed 1 year ago

Verified 8fe7ca48

server : hide ctx_sampling->prev behind API (#3696)

ggerganov committed 1 year ago

Verified 00ae55b3

Update readme to document multimodal in server

monatis committed 1 year ago

3d6a687f

server : minor style

ggerganov committed 1 year ago

Verified dd1af2ed

Merge branch 'server-rev' of https://github.com//ggerganov/llama.cpp into server-rev

monatis committed 1 year ago

a4d69d8b

Update readme to document multimodal in server

monatis committed 1 year ago

2679c432

server : completion requests remember slot_id

ggerganov committed 1 year ago

Verified a8063171

editorconfig : new line in index.html

ggerganov committed 1 year ago

Verified f305d643

Do not save/load image_data to localStorage

monatis committed 1 year ago

5359fb92

server : bug fix for prompt caching

ggerganov committed 1 year ago

Verified f67d9713

server : refactor ctx_sampling init + n_ctx + names

ggerganov committed 1 year ago

Verified 569ebf11

server : fix crash in Debug on macOS (I have no idea why this fixes it!?)

ggerganov committed 1 year ago

Verified ef18f4d5

server : fix switch fallthrough

ggerganov committed 1 year ago

Verified 197a0a9e

clip : link to ggml, not to llama

ggerganov committed 1 year ago

Verified 715f384a

make : silence stb warnings

ggerganov committed 1 year ago

Verified 4b4ab722

Merge branch 'master' into server-rev

ggerganov committed 1 year ago

Verified 176993c8

batched : add len CLI argument

ggerganov committed 1 year ago

Verified 22c69a27

fix image load + view image in chat

FSSRepo committed 1 year ago

2eb4c11e

server : fix multibyte handle in partial response (#3706)

jhen0409 committed 1 year ago

Verified 17b23eb9

CLBlast: Add outer loops over src0 for broadcasting in mulmat

shibe2 committed 1 year ago

465219b9

sampling : refactor init to use llama_sampling_params (#3696)

ggerganov committed 1 year ago

Verified d1031cf4

server : logs + minor code style

ggerganov committed 1 year ago

Verified 778c070d

server : no need for atomic int - already using mutex

ggerganov committed 1 year ago

Verified 5d540e80

server : bach has to be allocated for n_parallel sequences

ggerganov committed 1 year ago

Verified 113dd600

added thread safe pipeline

FSSRepo committed 1 year ago

6b2437e3

gguf : support big endian platform (#3552)

chenqiny committed 1 year ago

Verified 8cf19d60

server : fix uninitialized sampling context (close #3685)

ggerganov committed 1 year ago

Verified a0edf73b

ggml : fix rope + llama minor optimizations (#3560)

GermanAizek committed 1 year ago

Verified f439e506

Older