Add renamed "gpt-4-turbo-preview" model

• Add renamed "gpt-4-turbo-preview" model

Better image understanding + bugfixes

What's Changed

• better image understanding without gpt-4-vision using LLAVA-13b
• Fix conversation mentions bug
• Fix openai api Key url by @Paillat-dev in #462

Full Changelog: v12.3.6...v12.3.7

Make drawing understanding better

• Model understands how prompting works a bit better.

Midjourney prompt engineer, performance fixes

Full Changelog: v12.3.3...v12.3.5

• Fix more bug dumps in console from timeouts
• Add temperature and top_p parameter support for connected chats
• Update search to use the new models and make it a bit more stable and reliable
• Midjourney prompt engineer personality (opener file)
  

Make auto drawing disabled by default in /gpt converse

• Make auto drawing disabled by default in /gpt converse
  Full Changelog: v12.3.2...v12.3.3

Attempt to fix desync

Full Changelog: v12.3.1...v12.3.2

Drawing within /gpt converse!

• You can now work with GPT in a conversation to create images. This will work when you are in a gpt-vision conversation, the model will be able to draw (up to 4) images by itself, and will also be able to see images that you upload. Keep in mind that the vision API is limited to 100 requests per day, so you don't want to be spamming this until the API is fully released. This feature will ONLY work when the conversation model is gpt-4-vision-preview.

Full Changelog: v12.3.0...v12.3.1

Create images in /gpt converse!

• You can now work with GPT in a conversation to create images. This will work when you are in a gpt-vision conversation, the model will be able to draw (up to 4) images by itself, and will also be able to see images that you upload. Keep in mind that the vision API is limited to 100 requests per day, so you don't want to be spamming this until the API is fully released. This feature will ONLY work when the conversation model is gpt-4-vision-preview.

Full Changelog: v12.2.9...v12.3.0

Fix dalle-3 default model param

We were passing natural into the api call by default earlier, causing the default images to look not as dalle-3 is expected to look, as the actual default param is vivid

Fix max tokens issue for older models

What's Changed

• fixed gpt3.5 ask bug and bumped version by @rwywawn in #411

Full Changelog: v12.2.7...v12.2.8