An AI framework for building cool things.
LLMs are fantastic for knowledge generation and reasoning as they're pre-trained on large amounts of publicly available data.
In practice, we need LLMs to be able to reason about our own data. Fine-tuning them is often impractical. Embedding data into LLMs is a viable alternative, but requires domain expertise, and considerable effort to dealing with prompt limitations
Botastic aims to make it easy to embed your own data into LLMs.
Botastic also aims to provide out-of-the-box solutions to satisfy common use cases.
It would to be easily integrated into existing systems and scalable to meet business requirements.
We're building AI-based applications, we need Botastic to simplify the process.
Connect to OpenAI's LLM
Store and manage your own data as LLM's external brain
Restful APIs
Conversations management to help you build dialogue interface based applications
Bot/Avatar management and builtin useful bots: emotion recognition, sentiment analysis, spam checker, etc. Please refer to [TBD].
Customize Middlewares to extend Botastic's functionality
Store vectors in Postgres and other databases
Better strategies to split the data and build text chunks
Export data
Support other LLMs
Fine-tune own models
visit https://developers.pando.im/console/botastic to get your API key/secret, and create a new bot.
run the following command to talk to Botastic
curl --location --request POST 'https://botastic-api.pando.im/api/conversations/oneway' --header 'X-BOTASTIC-APPID: YOUR_BOTASTIC_APP_ID' --header 'Content-Type: application/json' --data-raw '{ "bot_id": YOUR_BOT_ID, "content": "How do you view the future of blockchain, and what benefits does it have for human being? Respond as short as possible like a Zen Master.", "category": "plain-text"}'
Please refer to Guide and API Reference for more details.
Ask Vitalik: A digital avatar of Vitalik Buterin.
Ask Scott Alexander: A digital avatar of Scott Alexander.
The author also wrote a blog post about how she built it.
Please refer to docs/install.md for manual installation.
TBD