Commit Graph
1479 Commits
Author SHA1 Message Date
Cohee 13db803bc6 CC: Remove auto-retry on code 429 2025-09-02 14:32:56 +03:00
fe8ebedeb2 fix: Resolve streaming output for Vertex AI Express Mode (#4474)
* fix(api): Resolve streaming output for Vertex AI Express Mode

* chore: Remove unused 'util' variable

* chore: Restore original debug log

* Even cleaner diff

---------

Co-authored-by: AliceRabbit <b290627396@gmail.com>
Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-09-01 16:49:55 +03:00
Cohee 285b09ceb7 NanoGPT: Add as captioning source 2025-08-31 19:00:27 +03:00
Cohee 5eea41d315 NanoGPT: Pull max context and vision capability from endpoint 2025-08-31 18:23:39 +03:00
Cohee 57cd001221 NanoGPT: Enable prompt cache for Claude models 2025-08-31 16:57:03 +03:00
Cohee bc2d3300df DeepSeek: Apply semi-strict PPP for reasoner 2025-08-31 16:45:57 +03:00
as-gh2andas-gh2 26a67b9afe Avoid consuming fetch response twice (#4460)
Fixes #4459

Co-authored-by: as-gh2 <->
2025-08-29 19:43:05 +03:00
Cohee 80fa7f96f1 Caption: Improve Ollama models handling 2025-08-27 00:45:24 +03:00
Cohee 5e4bb0415e OpenRouter: Add Google image output parsing 2025-08-26 22:01:38 +03:00
Cohee 0d84fa67bd gemini-2.5-flash-image-preview 2025-08-26 21:34:02 +03:00
Cohee 12ac171979 Temporarily disable cometapi provider 2025-08-25 20:41:26 +03:00
TensorNullandCohee 41c74fd142 Add CometAPI as a model provider in "Chat Completion" (#4402)
* feat: Add support for CometAPI, including API key management and model selection features.

* Some code fixes

* Fix loading model from presets

* Add logo image

* Enable tool calling, reasoning effort, reasoning display, structured generation

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-08-22 20:30:42 +03:00
sirius422andsirius422 4d9f229383 Add Variety Boost option for NovelAI image generation. (#4417)
Co-authored-by: sirius422 <sirius42@gfw.moe>
2025-08-22 20:27:17 +03:00
Cohee ad960197b7 Captioning: Load external model lists for OpenRouter, AIML and Pollinations 2025-08-21 22:09:13 +03:00
ershang-fireworksandCohee e9be0f1c64 Add Fireworks AI Provider Support (#4374)
* add fireworks provider

* fix

* add context length

* Fireworks fixes

* Add logo image

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-08-14 20:52:46 +03:00
Cohee 979fe60ebc Fix #4312 mostly functional chub import 2025-08-13 23:24:27 +03:00
buzzandCohee 0e5b509dc3 Drop Soulkyn import (#4372)
* Fix Soulkyn import

Generate and supply necessary headers to Soulkyn API request.
Fixes #4323

* Drop soulkyn import

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-08-13 22:25:21 +03:00
Cohee 69d256bd64 Prevent server crash when trying to backup settings of a deleted user 2025-08-09 17:09:58 +03:00
Cohee 5935aedb2e gpt5 thonking effort 2025-08-07 22:23:31 +03:00
Cohee deb417197d jippity 5 2025-08-07 21:52:10 +03:00
Ni-co-la-sandCohee 894f3e986e Add gpt-4o-mini-tts model with character voice instructions #3879 (#4352)
* Add gpt-4o-mini-tts model with character voice instructions

* Restyle instructions block

* Add debug log for tts request

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-08-06 21:52:21 +03:00
kallewoofandCohee c8612e2ab6 feature: ignore budget switch (#4354)
* feature: ignore budget switch

The budget cap default at 25% is a big gotcha at the start of chats, as a person may assume that a constant lore book entry will always be present, no matter what. This pull request adds a budget ignore switch which ignores the budget cap and includes the lore book entry even if it would exceed the budget cap.

* Move the toggle to entry details

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-08-06 21:30:42 +03:00
Cohee f1fcd68c99 WTF Anthropic??? 2025-08-06 08:35:01 +00:00
Cohee 4db44af110 Support gzip compression for tokenizer downloads (#4348)
* Support gzip compression for tokenizer downloads

* Use Uint8Array for buffer conversion

* Add gunzip support for manually downloaded tokenizer files

* Remove underlying buffer access

* Remove additional buffer allocations
2025-08-03 13:38:41 +03:00
Cohee 0e38bfbf05 Feat/moonshot api (#4330)
* moonshot

* Partial mode + JSON schema

* Add logo image

* Limit max temp to 1

* Add to captioning extension
2025-07-31 00:01:04 +03:00
Cohee 086f873f2f Deprecate 01.ai chat completion source (#4327) 2025-07-28 21:29:01 +03:00
Cohee 21d4c37846 Do not attempt to DNS resolve localhost
Fixes #4290
2025-07-27 21:42:45 +03:00
An5w1randCohee c283025b15 feat: add tts provider MiniMax (#4309)
* feat: add tts provider MiniMax

* fix: address code review feedback from @Cohee1207

- Moved inline styles to CSS classes in minimax.js
- Removed unused MINIMAX API key from secrets.js
- Removed static headers from debug logs
- Changed default API host to international endpoint
- Sorted TTS providers alphabetically in index.js

* feat: improve error handling and add language-specific test sentences based on selected voice

* fix: if no language mapping to MiniMax Format should use "auto"

* fix: move this minimax-tts.css under /public/scripts/extensions/tts/

* refactor: move minimax api communication to Node backend

* chore: unify import style for minimax router

* Move css into subfolder

* More secure secret key handling

* Better UI layout

* Finish secrets update

* Replace with a normal import

* This block shouldn't be here

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2025-07-27 15:56:26 +03:00
Coheeandgemini-code-assist[bot] 29815fbdda AI Studio/Vertex AI: add seed param (#4314)
* AI Studio/Vertex AI: add freq.pen, pres.pen and seed params

* Update src/endpoints/backends/chat-completions.js

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Add missing models to Google lists

* Do not send penalties to gemma and learnlm

* Disable penalties

* Fix not sending seed to Groq

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-07-27 15:10:19 +03:00
Cohee 7b67769c92 Fix convert to persona with unusual names (#4319)
* Fix convert to persona with unusual names

* Fix picking new persona on current persona deletion

* Prefer actual path on page reload
2025-07-27 14:16:15 +03:00
Cohee 2c0d8d71bc Browser launch: Add support for Brave browser 2025-07-26 18:29:15 +03:00
Cohee a8bb8a6703 Add hint popup on string length overflow 2025-07-26 18:20:45 +03:00
Cohee 82d9fa79e8 Add vectorization via Vertex AI (#4311)
* Add vectorization via Vertex AI

* Enable batching for Google vectors

* Add batch embeddings for Vertex

* Split embed methods for Vertex/AI Studio
2025-07-25 23:59:30 +03:00
Cohee fdc51a1211 Claude model deprecations (#4306)
* Claude: deprecate Claude 2 models

* Claude: flatten tool call schema

* Remove deprecated Sonnet 3 from captioning list
2025-07-25 22:06:12 +03:00
Cohee 29702a3f1b Merge pull request #4302 from cloudskytian/staging
Console access log output is now controlled by enableAccessLog
2025-07-25 22:02:27 +03:00
Cohee d641cbecc4 Remove Scale Spellbook from CC sources (#4293) 2025-07-24 22:02:45 +03:00
Cohee 38a5e4a4f2 Refactor BYAF parsing (#4303)
* Refactor BYAF parsing

* Refactor ByafParser to use private field for data and improve type annotations
2025-07-24 21:31:45 +03:00
cloudskytian d1a82c85a5 Console access log output is now controlled by enableAccessLog 2025-07-24 19:16:21 +08:00
Cohee 7edd81f183 Remove legacy key from config extras.promptExpansionModel 2025-07-24 00:58:12 +03:00
Cohee c627d77c46 Add toggle and user agent filter for CacheBuster middleware (#4301)
* Add toggle and user agent filter for CacheBuster middleware
Closes #4290 [BUG] SillyTavern takes ~45 seconds to load initially

* Use single instance of CacheBuster class

* Cache config values

* Consistent log header

* Remove unnecessary empty line at the beginning of byaf.js
2025-07-24 00:45:42 +03:00
CoheeandCopilot e0b020f211 Import BYAF (#4300)
* allow byaf import format

* Add byaf import converters

* Fix import of image and scenario

* Update src/byaf.js

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Log JSON parsing errors

* Use fsPromises for file acces

* Fix callback variable name

* Clear file input value after file upload to allow re-uploading

---------

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-07-23 23:05:14 +03:00
Cohee 7429726eb1 Remove Window AI from CC sources (#4294) 2025-07-22 19:45:54 +03:00
Cohee d070982ad7 Bump multer field size limit
Closes #4291
2025-07-21 20:26:18 +03:00
Coheeandgemini-code-assist[bot] 5d56992ea8 Feat/global install (#4289)
* Update npm dependencies

* Add global paths mode
Closes #4209

* Update src/command-line.js

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Async => Promise.resolve

* Remove unnecessary Promise.resolve()

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-07-21 00:24:49 +03:00
Cohee c292f64163 Add generation type triggers to world info (#4286)
* Add generation type triggers to world info

* Simplify includes check

* Refactor getEntryField validation and default value handling

* Remove invalid attribute

* Check for a valid trigger
2025-07-20 00:22:54 +03:00
Cohee e7ad36a758 Merge branch 'release' into staging 2025-07-17 13:49:35 +00:00
Cohee a64aae3edf Fix recover.js script 2025-07-17 13:49:15 +00:00
Cohee 7894e4d361 Google: Fix video inlining with PPP
Fixes #4268
2025-07-15 23:28:26 +03:00
Cohee 0e1d29acfe Bump request payload limits 2025-07-15 23:23:07 +03:00
Cohee f8f9ee8a0e NanoGPT: Add backend web search 2025-07-15 22:47:10 +03:00