Are there security best practices for each of these tools or is that too early?

“our team was blown away by how much Codex was able to accelerate its own development.”

At what point will LLMs be autonomously self creating new versions of themselves?

To anyone trying this, does this unlock anything you tried to do with the past LLM models but failed and now you can try again? Do you find this as an incremental improvement or something that brings in new opportunities?

What's the cheapest device specs that this could realistically run on?

I haven't quite figured out if the open weights they released on huggingface amount to being able to run the (realtime) model locally - i hope so though! For the larger model with diarization I don't think they open sourced anything.

The HF page suggests yes, with vllm.

> We've worked hand-in-hand with the vLLM team to have production-grade support for Voxtral Mini 4B Realtime 2602 with vLLM. Special thanks goes out to Joshua Deng, Yu Luo, Chen Zhang, Nick Hill, Nicolò Lucchesi, Roger Wang, and Cyrus Leung for the amazing work and help on building a production-ready audio streaming and realtime system in vLLM.

https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-26...

https://docs.vllm.ai/en/latest/serving/openai_compatible_ser...

I’ve been considering building something similar for a while. It is super interesting to see this implemented

Do you have a library of similar prompts you could link here?

I expect QA and AI hand-holding roles to increase, but I don’t see that shift happening yet.

In my opinion, the way this will play out is with a significant amount of validation and human oversight to fully utilize these LLMs. As you mentioned, I recommend giving the AI room for error and improving the experience of manually checking everything. Maybe create a tool to facilitate manually checking the output?

This is a valuable read: https://www.ufried.com/blog/ironies_of_ai_1/

what are you current go-to tools to speed up the review?

This is awesome. I love the documentation.

Docs created by Mintlify https://www.mintlify.com/

Yep, thanks Mintlify!

This is pretty neat. All serious products get a lot of questions about prompt vulnerabilities that this would address.