The commit updates AI model settings for GPT-3.5 by allowing up to 4000 tokens for context size. It also adds a new option for GPT-3.5 Turbo 16k with a maximum context of 16000 tokens. Additionally, the commit limits the context size when exceeding the maximum limit for each model.
# PR Checklist
- [ ] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [x] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [x] Did you added a type def?
# Description
- Screenreader readable "Show Help" button.
Also modified checkbox
# PR Checklist
- [ ] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [ ] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [ ] Did you added a type def?
# Description
# PR Checklist
- [YES] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [YES] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [NO] Did you added a type def?
# Description
# PR Checklist
- [v] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [v] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [NO] Did you added a type def?
# Description
### Bug
send message double time by enter when writing korean.
This issue occurs when input's composition is not end.
https://github.com/kwaroran/RisuAI/assets/60877502/758b3168-d63c-4b64-8ca0-6a5afefad749
### fix
block send message when event is composing.
# PR Checklist
- [ ] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [ ] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [ ] Did you added a type def?
# Description
# PR Checklist
- [ O ] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [ O ] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [ O ] Did you added a type def?
# Description
- This commit adds new entries in the language files and UI support for
auto-translate messages.
- The new feature uses a textarea input field to translate messages as
they are typed, and the translated text is shown in a new field. Also,
the language settings now provide an option to enable/disable auto
message translation.
- The code changes include adding a new function to update the input
sizes for both input fields and some minor code refactoring to support
new functionality.
- This commit adds the import for the ParseMarkdown function from the
parser module required for translating messages in the Suggestion chat
screen.
- The code changes also include adding a check for the translator
variable in the button toggle handling to avoid unwanted button display.
Lastly, the commit updates the template for rendering translated suggest
messages.
- Improve user prompt for AutoSuggest
# PR Checklist
- [YES] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [YES] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [NO] Did you added a type def?
# Description
# PR Checklist
- [ ] Did you check if it works normally in all models? *ignore this
when it dosen't uses models*
- [ ] Did you check if it works normally in all of web, local and node
hosted versions? if it dosen't, did you blocked it in those versions?
- [ ] Did you added a type def?
# Description
