by Robert Sarvis | Dec 24, 2025 | Featured, Insights |
The Results Let me start with the punchline: Metric Base Phi-3-Mini After 60 Hours Training Delta Pass@10 100% 0% -100% Overall Pass Rate 60% 0% -60% Total Solutions 18/30 0/30 -18% I did not just fail to improve the model. I took a working Go code generator and... by Robert Sarvis | Sep 10, 2025 | Insights |
A few years ago, I had a conversation with an AWS Solutions Architect who recommended Claude from Anthropic. Since then, Claude has been my go-to model for AI code completion. I’ve tested new offerings as they’ve been released, but I always ended up back with Claude....