Hacker News new | ask | show | jobs
Shape the Future of AI: Llama Users, We Need Your Feedback at Meta!
6 points by diegoguerra96 638 days ago
I work in the team at Meta building Llama models. We are looking to better understand how we can improve Llama models. We want to understand your challenges to build open source AI tools that help you the most. Your input is informing our roadmap and how we support the community to help accelerate adoption.

Your feedback will be invaluable in shaping our product roadmap and potentially guiding the content of these community support initiatives, including possibly exploring hosting workshops focused on common Llama use cases and challenges.

Here's what we'd love to understand:

If you're using any models:

1. Most important considerations for LLM selection?

2. What models are you using in production? If not Llama, why not?

If you're using Llama:

3. Use case description for Llama?

4. Llama Performance: Strengths? Weaknesses? Improvement areas?

5. Feature requests?

6. Deployment challenges with Llama?

7. AI stack: RAG, fine-tuning, agentic flows?

8. Data types you're working with?

9. Model fine-tuning challenges?

10. Gaps in developer tools?

1 comments

I tested Llama 3.2 on my MacBook, and it runs impressively fast. I plan to look more deeply into its capabilities and have some specific use cases in mind for further testing. Fine-tuning could be improved, and better integration with AI stacks like RAG would be helpful. Overall, it's a solid model with great potential! https://x.com/mustafaergisi/status/1839647877498278311
Can you explain how you would like fine tuning to be improved?
more options for customizable tuning would be great.