Commit graph

  • 23d3f6909a
    Update README.md oobabooga 2023-05-11 10:21:20 -03:00
  • 400f3648f4
    Update docs/README.md oobabooga 2023-05-11 10:10:24 -03:00
  • 2930e5a895
    Update README.md oobabooga 2023-05-11 10:04:38 -03:00
  • 0ff38c994e
    Update README.md oobabooga 2023-05-11 09:58:58 -03:00
  • e6959a5d9a
    Update README.md oobabooga 2023-05-11 09:54:22 -03:00
  • dcfd09b61e
    Update README.md oobabooga 2023-05-11 09:49:57 -03:00
  • ac9a86a16c
    Update llama.cpp-models.md oobabooga 2023-05-11 09:47:36 -03:00
  • 7a49ceab29
    Update README.md oobabooga 2023-05-11 09:42:39 -03:00
  • 943b5e5f80 Minor bug fix oobabooga 2023-05-10 23:54:25 -03:00
  • 9695bfe117 Add an option for cleaning up html in superbooga oobabooga 2023-05-10 23:23:37 -03:00
  • c7ba2d4f3f Change a message in download-model.py oobabooga 2023-05-10 19:00:14 -03:00
  • 1309cdd257
    Add a space oobabooga 2023-05-10 18:03:12 -03:00
  • 3e19733d35
    Remove obsolete comment oobabooga 2023-05-10 18:01:04 -03:00
  • 4ab5deeea0
    Update INSTRUCTIONS.TXT oobabooga 2023-05-10 18:00:37 -03:00
  • d7d3f7f31c
    Add a "CMD_FLAGS" variable oobabooga 2023-05-10 17:54:12 -03:00
  • b7a589afc8 Improve the Metharme prompt oobabooga 2023-05-10 16:09:32 -03:00
  • e5b1547849 Fix reload model button oobabooga 2023-05-10 14:43:28 -03:00
  • b01c4884cb Better stopping strings for instruct mode oobabooga 2023-05-10 14:22:38 -03:00
  • 6a4783afc7 Add markdown table rendering oobabooga 2023-05-10 13:41:23 -03:00
  • 57dc44a995
    Update README.md oobabooga 2023-05-10 12:48:25 -03:00
  • f5592781e5
    Update README.md oobabooga 2023-05-10 12:19:56 -03:00
  • f1d10edcb7
    Update README.md oobabooga 2023-05-10 12:13:14 -03:00
  • 181b102521
    Update README.md oobabooga 2023-05-10 12:09:47 -03:00
  • 3316e33d14 Remove unused code oobabooga 2023-05-10 11:59:59 -03:00
  • ec14d9b725
    Fix custom_generate_chat_prompt (#1965) Alexander Dibrov 2023-05-10 09:29:59 -05:00
  • 32481ec4d6 Fix prompt order in the dropdown oobabooga 2023-05-10 02:24:09 -03:00
  • dfd9ba3e90 Remove duplicate code oobabooga 2023-05-10 02:07:22 -03:00
  • cd36b8f739 Remove space oobabooga 2023-05-10 01:41:33 -03:00
  • c35860ff2f Add a link to silero samples oobabooga 2023-05-10 01:39:35 -03:00
  • bdf1274b5d Remove duplicate code oobabooga 2023-05-10 01:34:04 -03:00
  • ba445cf59f Fix some galactica templates oobabooga 2023-05-09 22:58:59 -03:00
  • 3b1de7e8bc Remove redundant presets oobabooga 2023-05-09 22:56:19 -03:00
  • 3913155c1f
    Style improvements (#1957) oobabooga 2023-05-09 22:49:39 -03:00
  • 334486f527
    Added instruct-following template for Metharme (#1679) minipasila 2023-05-10 04:29:22 +03:00
  • 1aaa47070a
    Expand Open Assistant support (#1735) Carl Kenner 2023-05-10 09:10:29 +09:30
  • 814f754451
    Support for MPT, INCITE, WizardLM, StableLM, Galactica, Vicuna, Guanaco, and Baize instruction following (#1596) Carl Kenner 2023-05-10 09:07:31 +09:30
  • 06c7db017d
    Add config for pygmalion-7b and metharme-7b (#1887) Matthew McAllister 2023-05-09 16:31:27 -07:00
  • fe4dfc647d
    SileroTTS preview (#1934) missionfloyd 2023-05-09 17:28:59 -06:00
  • 8fa5f651d6 Style changes oobabooga 2023-05-09 20:20:35 -03:00
  • e9e75a9ec7
    Generalize multimodality (llava/minigpt4 7b and 13b now supported) (#1741) Wojtab 2023-05-10 01:18:02 +02:00
  • a2b25322f0
    Fix trust_remote_code in wrong location (#1953) Wesley Pyburn 2023-05-10 00:22:10 +02:00
  • 13e7ebfc77 Change a comment oobabooga 2023-05-09 15:56:32 -03:00
  • 218bd64bd1
    Add the option to not automatically load the selected model (#1762) LaaZa 2023-05-09 18:52:35 +00:00
  • b8cfc20e58
    Don't install superbooga by default oobabooga 2023-05-09 14:17:08 -03:00
  • cf6caf1830
    Make the RWKV model cache the RNN state between messages (#1354) Maks 2023-05-09 16:12:53 +02:00
  • 641500dcb9
    Use current input for Impersonate (continue impersonate feature) (#1147) Kamil Szurant 2023-05-09 07:37:42 +02:00
  • a5bb278631
    Bump accelerate from 0.18.0 to 0.19.0 (#1925) dependabot[bot] 2023-05-09 02:17:27 -03:00
  • 29727c6502
    Fix Windows PATH fix (#57) jllllll 2023-05-08 23:49:27 -05:00
  • 020fe7b50b
    Remove mutable defaults from function signature. (#1663) IJumpAround 2023-05-08 21:55:41 -04:00
  • 32ad47c898
    added note about build essentials to WSL docs (#1859) shadownetdev1 2023-05-08 21:32:41 -04:00
  • 791a38bad1
    [extensions/openai] Support undocumented base64 'encoding_format' param for compatibility with official OpenAI client (#1876) Jeffrey Lin 2023-05-08 18:31:34 -07:00
  • d78b04f0b4
    Add error message when GPTQ-for-LLaMa import fails (#1871) Matthew McAllister 2023-05-08 18:29:09 -07:00
  • 99d2dd8d0d
    Fix - Ignore OS files from source control (#1911) Kévin Legrand 2023-05-09 03:17:37 +02:00
  • 0e27b660e8
    Fix issue where n_results/k > index size (#1929) kaiokendev 2023-05-08 20:16:00 -04:00
  • 68dcbc7ebd Fix chat history handling in instruct mode oobabooga 2023-05-08 16:41:21 -03:00
  • 79ac94cc2f
    fixed LoRA loading issue (#1865) Clay Shoaf 2023-05-08 15:21:55 -04:00
  • 8818967d37
    Fix typo in docs/Training-LoRAs.md (#1921) Arseni Lapunov 2023-05-08 20:12:39 +02:00
  • b5260b24f1
    Add support for custom chat styles (#1917) oobabooga 2023-05-08 12:35:03 -03:00
  • b040b4110d Bump llama-cpp-python version oobabooga 2023-05-08 00:21:17 -03:00
  • 63898c09ac Document superbooga oobabooga 2023-05-08 00:11:31 -03:00
  • 82eb284a3f Fix a bug in silero tts oobabooga 2023-05-07 23:59:43 -03:00
  • 0c048252b5
    Fix character menu when default chat mode is 'instruct' (#1873) Matthew McAllister 2023-05-07 19:50:38 -07:00
  • d3ea70f453
    More trust_remote_code=trust_remote_code (#1899) EgrorBs 2023-05-08 05:48:20 +03:00
  • 56a5969658
    Improve the separation between instruct/chat modes (#1896) oobabooga 2023-05-07 23:47:02 -03:00
  • 9754d6a811 Fix an error message oobabooga 2023-05-07 17:44:05 -03:00
  • ba65a48ec8
    trust_remote_code=shared.args.trust_remote_code (#1891) camenduru 2023-05-07 23:42:44 +03:00
  • b3bbda22d1 Exception handling oobabooga 2023-05-07 16:50:45 -03:00
  • 8515f19078 Fix typing error oobabooga 2023-05-07 16:25:39 -03:00
  • 6b67cb6611 Generalize superbooga to chat mode oobabooga 2023-05-07 15:01:14 -03:00
  • ec1cda0e1f Rename superbig to superbooga oobabooga 2023-05-07 13:29:49 -03:00
  • 8216c8f732 Handle multiline inputs oobabooga 2023-05-07 11:54:26 -03:00
  • a35a2fab02 Reorganize superbig ui oobabooga 2023-05-07 11:30:16 -03:00
  • befa307c42 Add support for loading multiple URLs into superbig oobabooga 2023-05-07 11:07:16 -03:00
  • 04eca9b65b Make chunk length/count customizable oobabooga 2023-05-07 05:02:04 -03:00
  • 8c06eeaf84 Add requirement oobabooga 2023-05-07 04:50:12 -03:00
  • 58f9a30be1 Add support for input file/url in superbig oobabooga 2023-05-07 04:49:02 -03:00
  • 6afba065a1 Only update the chroma database when necessary oobabooga 2023-05-07 04:03:17 -03:00
  • 5a4bd3918c
    Add SuperBIG extension (alpha) (#1548) kaiokendev 2023-05-07 02:50:12 -04:00
  • 81be7c2dd4 Specify gradio_client version oobabooga 2023-05-06 21:50:04 -03:00
  • 85238de421 Remove unused variable oobabooga 2023-05-06 11:03:12 -03:00
  • de9c4e260e
    Minor fixes to elevenlabs_tts oobabooga 2023-05-06 10:57:34 -03:00
  • b03a2ac512
    Elevenlabs Extension Improvement and migration to official API (#1830) Steve Randall 2023-05-06 14:56:31 +01:00
  • 126d216384
    Fix possible crash (#53) Semjon Kravtšenko 2023-05-06 07:14:09 +03:00
  • 56f6b7052a Sort dropdowns numerically oobabooga 2023-05-05 23:14:56 -03:00
  • ee3c8a893e
    Update Extensions.md oobabooga 2023-05-05 19:04:50 -03:00
  • 8aafb1f796
    Refactor text_generation.py, add support for custom generation functions (#1817) oobabooga 2023-05-05 18:53:03 -03:00
  • 876fbb97c0
    Allow downloading model from HF branch via UI (#1662) Tom Jobbins 2023-05-05 17:59:01 +01:00
  • 849ad04c96 Change background color of instruct code blocks oobabooga 2023-05-05 12:02:45 -03:00
  • c728f2b5f0 Better handle new line characters in code blocks oobabooga 2023-05-05 11:22:36 -03:00
  • 207a031e8d CSS change to instruct mode oobabooga 2023-05-05 00:36:15 -03:00
  • e5d6d822b1 Minor CSS change to instruct mode oobabooga 2023-05-04 23:41:00 -03:00
  • a50c2ab82a Add missing file oobabooga 2023-05-04 23:29:46 -03:00
  • 00e333d790 Add MOSS support oobabooga 2023-05-04 23:19:23 -03:00
  • f673f4a4ca Change --verbose behavior oobabooga 2023-05-04 15:56:06 -03:00
  • 97a6a50d98 Use oasst tokenizer instead of universal tokenizer oobabooga 2023-05-04 15:55:39 -03:00
  • b6ff138084 Add --checkpoint argument for GPTQ oobabooga 2023-05-04 15:17:20 -03:00
  • dbddedca3f Detect oasst-sft-6-llama-30b oobabooga 2023-05-04 15:13:37 -03:00
  • 1436c5845a
    fix ggml detection regex in model downloader (#1779) Wojtek Kowaluk 2023-05-04 16:48:36 +02:00
  • bd531c2dc2
    Make --trust-remote-code work for all models (#1772) Mylo 2023-05-04 07:01:28 +02:00
  • 0e6d17304a Clearer syntax for instruction-following characters oobabooga 2023-05-03 22:50:39 -03:00