Our great sponsors
-
guidance
Discontinued A guidance language for controlling large language models. [Moved to: https://github.com/guidance-ai/guidance] (by microsoft)
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
-
agnai
AI Agnostic (Multi-user and Multi-bot) Chat with Fictional Characters. Designed with scale in mind.
I am fond of Microsoft Guidance: https://github.com/microsoft/guidance/ which allows you to have highly structured conversations with many different LLMs. Your Python code can jump in and modify the conversation at any time, allowing your program to easily inject tool usage into the conversational chain. I recommend reading each example very carefully.
There is also a fork to allow easy use with llama.cpp: https://github.com/Maximilian-Winter/guidance (use his matching llama-cpp-python fork as well). You can also use it with GPTQ for GPU-accelerated GPTQ-quantized models.
For multiple characters, there have been tries out there so you can try copying from them (I practically am almost clueless about this topic, here is one of these tries) or you could also try making your own idea.
Related posts
- DataFrameAndNotebooksAmsterdam2024 – Discovering why trains come in late in NL
- Why Vector Compression Matters
- Scalable Load Balancing Having Cloud GPU Service Salad Tutorial With Whisper Transcriber Gradio APP
- Show HN: I made a website that converts YT videos into step-by-step guides
- Metrics for bias in machine learning datasets