ochafik/llama.cpp

Pull Requests Commits

reinstate trailing \n

ochafik committed 1 year ago

e41a5403

`server`: fix disconnection logic in test (before post response headers)

ochafik committed 1 year ago

52c5a624

`server`: use (new) Request::is_alive as set_content_provider called after status / headers sent

ochafik committed 1 year ago

6f693f14

Delete tests/.DS_Store

ochafik committed 1 year ago

2dc708c7

`server`: avoid calling sink.is_alive() after it died 🧟‍♂️

ochafik committed 1 year ago

6297d9fe

`server`: mime nit

ochafik committed 1 year ago

16ff5022

`server`: support cancellation of prompt processing

ochafik committed 1 year ago

03efb92f

`server`: introduce supposedly lighterweight is_alive in httplib (https://github.com/yhirose/cpp-httplib/pull/1956)

ochafik committed 1 year ago

42f54650

server: fix error status

ochafik committed 1 year ago

43e306e0

`server`: speed up cancel test setup

ochafik committed 1 year ago

d07387ca

`server`: update log

ochafik committed 1 year ago

0e9c4bf5

Update cancel.feature

ochafik committed 1 year ago

c5a0d57e

`server`: fix seed in tests (comma creates a tuple)

ochafik committed 1 year ago

231a5e49

`server`: fix cancel tests

ochafik committed 1 year ago

3f96ab04

`server`: simplify handle_tasks signature

ochafik committed 1 year ago

88c9b549

`server`: rm superfluous is_alive check in streamed code

ochafik committed 1 year ago

419e9952

`server`: test request cancellation (WIP)

ochafik committed 1 year ago

5f00747a

`tests`: allow artificial slowdown of sampling for tests

ochafik committed 1 year ago

4dcb3ea9

`server`: support cancelling non-streamed requests

ochafik committed 1 year ago

1da67a39

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

readme : add tool (#9655)

akx committed 1 year ago

Verified 43bcdd97

ggml : add run-time detection of neon, i8mm and sve (#9331)

eddnjjn committed 1 year ago

Verified 6a0f7794

Enable use to the rebar feature to upload buffers to the device. (#9251)

mtavenrath committed 1 year ago

Verified 89f99449

readme : update hot topics

ggerganov committed 1 year ago

Verified b5de3b74

cmake : add option for common library (#9661)

iboB committed 1 year ago

Verified 44f59b43

[SYCL] add missed dll file in package (#9577)

NeoZhangJianyu committed 1 year ago

Verified 95bc82fb

Older