comparing AI models for research
Members compare Claude, Sonnet, and Gemini, debating strengths in coding, cost efficiency, and conceptual math research versus basic search.
- Sonnet 5.5 is highlighted as an exceptionally cost-effective and powerful model capable of handling the bulk of general workloads
- Gemini accessed through model providers is described as a solid, budget-friendly option for work tasks
- Disagreement surrounds Gemini, with some viewing it merely as a glorified search engine that struggles at math and coding
- Gemini is defended for academic research, excelling at finding sources and explaining conceptual mathematics and chemistry


