feat: add image text model provider separation and fal.ai integration#650

Merged

lalalune merged 13 commits intoelizaOS:mainfrom yoniebans:feat/add_image_text_model_provider_seperation_and_falai

Nov 29, 2024

+129-12

Commits on Nov 28, 2024

updated Characters.ts in core/types to include an optional imageModelProvider allowing for use of different models between textGeneration and imageGeneration. If imageModelProvider is not set, it d…
yoniebans
committed
given the option for distinction between the model being used for generateText and generateImage, added a new property for the imageModelProvider on the runtime class. This property is set to the i…
yoniebans
committed
updateds core models to include a new entry for fal.ai
yoniebans
committed
updated imageGeneration action validate function to include check for FAL_API_KEY not that it can be used for generateImage
yoniebans
committed
update core packages to include fal.ai client
yoniebans
committed
include image generation plugin if FAL_API_KEY is present by default
yoniebans
committed
updated runtime interfact to include property for imageModelProvider
yoniebans
committed
updated generation to include implementation for fal.ai image generation and leverage the new distinction between imageModelProvider and modelProvider. Have changed the checks from runtime.characte…
yoniebans
committed
updated plugin-image-generation environment variable checks to include fal.ai
yoniebans
committed
Merge branch 'main' into feat/add_image_text_model_provider_seperation_and_falai
yoniebans
committed
updated example env file for optional fal.ai vars
yoniebans
committed
added HEURIST and OPENAI as valid options for automatically enabling image generation
yoniebans
committed

Commits on Nov 29, 2024

Merge branch 'main' into feat/add_image_text_model_provider_seperation_and_falai
yoniebans
authored