Dev Rant: People 'Killing It' Use the Biggest Models with Most Thinking; Small Models are Just a Show
rickasaurus · x · 2026-08-12
Developer @rickasaurus shared a blunt observation: everyone he knows who is highly successful is using the largest available models with maximum thinking capabilities.
He directly called out those showcasing development with smaller or cheaper models like DeepSeek, claiming they are just putting on a show. This sparked a discussion about model selection versus cost-efficiency in real-world development.
More from Models
- Analyst Cancels Gemini Over Poor Quality, Warns Google Losing SOTA Moat — firstadopter · 2026-08-12
- Gemma 4 QAT Shows Significant Improvement in KV Cache Quantization, KLD Benchmarks Reveal — Anbeeld · 2026-08-12
- New OEIS Open Benchmark: Opus 4.8 Solves 30% of Unsolved Math Conjectures — xeophon · 2026-08-12
- xAI Releases Grok 4.6 Model API — legit_api · 2026-08-12
- DeepSeek + Local 3B Vision Model Powers Cloud-Free Agent Browsing — helloiamleonie · 2026-08-12
- DeepSeek V4 Pro General Availability Rolls Out, Testing Chinese Labs' Post-Training Prowess — zephyr_z9 · 2026-08-12