Experiences with local models for coding
Birgitta Böckeler shares her recent experiences with local LLMs for coding in an article on Martin Fowler's site. She evaluates several models by comparing them on two standard coding tasks, then selects the most promising one for daily use. The article details her methodology and findings, offering insights into the viability of local models for development work. No specific model names, versions, or benchmark numbers are mentioned in the excerpt, but the focus is on practical, hands-on assessment rather than theoretical comparison.
// why it matters
Provides practical insights into using local LLMs for coding tasks.