945 Commits

Author SHA1 Message Date
AlpinDale
62fed2ff6a fix: pass model parameter for aphrodite tokenizer 2024-08-24 22:43:20 +00:00
Cohee
3f026aea15 SD: Add prompt expansion warning. Disable quantization 2024-08-23 19:57:25 +03:00
Cohee
06e3d5f8de Rename MakerSuite => AI Studio 2024-08-21 21:00:17 +03:00
Alex Yancey
135ba2336e Hugging Face inference API for image generation 2024-08-19 23:02:15 -07:00
Cohee
ded26f9783 Fix Google translate result decoding
Inspired by #2522
2024-08-18 14:55:17 +03:00
Cohee
c2057da348 Move transformers.js model cache under the data root 2024-08-16 11:59:41 +00:00
Cohee
e707def7dd Add tokenizer for Gemma/Gemini 2024-08-16 00:00:43 +03:00
Cohee
7322dd1954 Add optional Claude system prompt cache. 2024-08-15 21:25:08 +03:00
Cohee
0869270c26 Increase Anthropic multimodal caption max_tokens 2024-08-15 21:12:01 +03:00
Cohee
cb7185fa12 [chore] Fix grammar, add JSDocs 2024-08-15 20:29:17 +03:00
Cohee
a08b3ec7fc Sort gallery images by date 2024-08-15 00:40:08 +03:00
Cohee
bc7ac12317 New ChatGPT model 2024-08-14 20:31:25 +03:00
Cohee
1883095e39 Move request to backend 2024-08-13 20:29:33 +03:00
Cohee
f305ba7ce7 Configurable ollama keep_alive
Closes #1859
2024-08-11 17:32:31 +03:00
Cohee
8df27254fd
Merge pull request #2640 from edk208/release
Adding Block Entropy API endpoints for chat, image, and video generation
2024-08-11 16:19:14 +03:00
Edward Kim
7546030c48 Removed the scheduler, unused and auto added by vs code by mistake 2024-08-10 14:11:34 -04:00
Edward Kim
59c66353a9 fixed linting spacing issue 2024-08-10 13:12:57 -04:00
Edward Kim
08d5a2826f fixed generation endpoints, added animation via gifs 2024-08-10 13:07:52 -04:00
Edward Kim
b9857eb315 Added block entropy endpoints for chat and image 2024-08-08 00:08:03 -04:00
Cohee
f3734ae285 Remove token limit from multimodal caption request 2024-08-06 22:45:05 +03:00
Alex Yancey
ace6e9d18d Fix incorrect JanitorAI/JannyAI import due to incorrect HTTP header 2024-08-03 05:29:33 -07:00
Succubyss
366070adee ultra re-addition (just in case), fixed system prompt line 2024-08-02 15:26:15 -05:00
Cohee
4b61bcc741
Merge pull request #2600 from Succubyss/gemini-1.5-pro-exp-0801 2024-08-02 21:39:19 +03:00
Succubyss
3f6e4fabb8 gemini updates 2024-08-02 13:30:54 -05:00
Cohee
9cd59d2e4b Pass app headers to featherless 2024-08-01 23:03:32 +03:00
Cohee
9deeca3eeb Pass min_p to ollama backend 2024-08-01 02:45:45 +03:00
Cohee
40ee236ca8 Add multimodal captioning for 01.ai 2024-08-01 01:34:49 +03:00
Cohee
cacd570af3 More supported samplers for Together 2024-07-28 21:56:33 +03:00
Cohee
5ad433c576 #2557 Put MistralAI prefix under a feature toggle 2024-07-27 19:57:40 +03:00
Cohee
ef59e8abae #2557 Improve Mistral prompt conversion 2024-07-25 21:04:57 +03:00
Cohee
5f2a73ac9f Expose "Allow fallback providers" for OpenRouter 2024-07-19 23:34:16 +03:00
blackmesataiwan
09b08a49d3 Update language code for Chinese translations(zh-TW)
The code changes in `translate.js` update the language code for Chinese translations. The `zh-TW` language code is changed to `zt`. This ensures consistency and compatibility with the translation system(https://libretranslate.com/languages).
2024-07-19 19:44:00 +08:00
Cohee
52497ea96d Interrupt Comfy gens on cancel 2024-07-19 00:39:21 +03:00
Cohee
7e4abaabff Interrupt AUTO1111 gens on cancel 2024-07-19 00:01:24 +03:00
Cohee
1effb66fd6 Add cancelling of SD gens 2024-07-18 23:23:33 +03:00
Cohee
a2c9e70082 Revert WASM worker threading 2024-07-17 15:02:27 +00:00
Cohee
c3d04066bc Allow WASM threading for non-Android. Fix eslint for transformers.mjs 2024-07-17 12:48:59 +00:00
Cohee
921297f1e5 Fix stats init for new users 2024-07-17 11:19:22 +00:00
Cohee
3fb83e47ec Run eslint 2024-07-14 14:07:23 +03:00
kwaroran
fe0b29f606 move group_only_greetings to characters.js 2024-07-12 08:36:35 +09:00
kwaroran
c3185d34c7 Add v3 writing and decorators 2024-07-11 15:11:35 +09:00
Risenafis
de7de03481
set transcribe 2024-07-10 00:57:09 +09:00
Risenafis
968eb3790f
fix transformers model changes 2024-07-09 03:09:42 +09:00
Cohee
f7e2d72ac7 Fix eslint 2024-07-07 22:40:03 +03:00
kingbri
0672c8422e Textgen: Add permissions check for TabbyAPI keys
There's no formal permissions checking in ST's UI, so add a temporary
check in the server endpoint before requesting a download.

Signed-off-by: kingbri <bdashore3@proton.me>
2024-07-07 14:40:24 -04:00
kingbri
69077f6a6e Textgen: Add model downloading for TabbyAPI
Tabby has a HuggingFace downloader API endpoint. Add direct support
in SillyTavern by using the same method as ollama, but with the correct
parameters exposed.

Signed-off-by: kingbri <bdashore3@proton.me>
2024-07-07 14:04:04 -04:00
Cohee
6bdbe7299f Merge branch 'staging' into vllm-tc 2024-07-06 11:18:38 +03:00
Cohee
54dea56a77 Selectable VAE for AUTO1111 2024-07-05 00:59:34 +03:00
Cohee
b682519042 Hide sampling method select from unsupported backends 2024-07-04 23:00:33 +03:00
Cohee
5da3921907 Make generation actually work 2024-07-04 22:51:47 +03:00