[QUESTION] How do we pick the best models

[markj]

🤔 What do you want to know about?

With almost weekly major changes in the model space.. how do we sensibly evaluate which one to be using for certain tasks without always chasing the next shiny thing.

📚 Context

Like all tools, change comes at a cost as well as a benefit. New models bring much more powerful features but if things are changing does that reduce effectiveness from previous learnings.

🎯 What would help you?

• Tutorial/How-To Guide
• Code Examples
• Links to Resources
• Best Practices
• Tool Recommendations
• Other (please specify)