ochafik/llama.cpp

Pull Requests Commits

Olivier Chafik committed 1 year ago

34f54dd1

Disable chat_completion tests of non-tool jinja mode

Olivier Chafik committed 1 year ago

729d2d36

try and avoid weird server test failure (spillage / parallelism between completion & tool call tests?)

Olivier Chafik committed 1 year ago

3bd6abeb

force printing </tool_call> on hermes 2 model if/as it's a special token

Olivier Chafik committed 1 year ago

1029ff90

test: leave model_hf_file blank

ngxson committed 1 year ago

5add261a

log prompt + nits

Olivier Chafik committed 1 year ago

82052466

Format test-chat.cpp

Olivier Chafik committed 1 year ago

f223df02

add llama_sampler_init_grammar_lazy instead of renaming the non-lazy

Olivier Chafik committed 1 year ago

5a64af6c

deprecate llama_sampler_init_grammar -> llama_sampler_grammar_init

Olivier Chafik committed 1 year ago

7d59bf44

nit: fix py import

Olivier Chafik committed 1 year ago

2bb3fed3

Update scripts/fetch_server_test_models.py to new compact hf_repo syntax + switch Hermes models

Olivier Chafik committed 1 year ago

96850432

Update test_chat_completion.py

Olivier Chafik committed 1 year ago

0c171f54

Update test_chat_completion.py

Olivier Chafik committed 1 year ago

06c4ca56

Fix debug + verbose

Olivier Chafik committed 1 year ago

3dcde9ea

simplify handle_apply_template

ngxson committed 1 year ago

c88f4a79

code style changes on test

ngxson committed 1 year ago

2d51c459

Merge remote-tracking branch 'origin/master' into tool-call

Olivier Chafik committed 1 year ago

8ef37a3c

sync: minja (#11499)

ochafik committed 1 year ago

Verified 3d804dec

vocab : correctly identify LF token for GPT-2 style BPE tokenizer (#11496)

mgroeber9110 committed 1 year ago

Verified ffd0821c

server : use lambda instead of std::bind (#11507)

danbev committed 1 year ago

Verified 4314e56c

server : (docs) added response format for /apply-template [no ci] (#11503)

isaac-mcfadyen committed 1 year ago

Verified 496e5bf4

readme : reference examples relative links (#11505)

guspan-tanadi committed 1 year ago

Verified 7919256c

increase http timeout to 12

ochafik committed 1 year ago

9591af1f

llama 3.2 1b now fails the weather tool call?

ochafik committed 1 year ago

7635912f

rm unused llama_param

ochafik committed 1 year ago

b831a6e0

server : update json snippets in README.md [no ci] (#11492)

danbev committed 1 year ago

Verified e0449763

debug logs are back

ochafik committed 1 year ago

18450e69

ochafik committed 1 year ago

81547e6f

split chat handler vs. parser around enum again

ochafik committed 1 year ago

f8e14bff

Update tests readme + add raw output to verbose log

ochafik committed 1 year ago

590c9793

Older