Google’s most intelligent workhorse model - the third Flash release of summer 2026 (3.5 Flash May, 3.6 Flash Jul, 3.7 Flash Aug 13). 1M context, 64K output, multimodal (text, image, video, audio, PDF) in, text out. Customizable thinking; tool use, search, and computer use.
Pricing: introductory $0.75/1M input, $3.75/1M output through Dec 31, 2026, then $1.50/$7.50. Batch 50% off.
Agents on Rails benchmark (Aug 2026, Le Mans round). 71.4% accuracy on 63 runs at $0.283 mean cost - the strongest accuracy-per-dollar below the top cluster, edging out Luna’s $0.014 cost-per-point tradeoff for mid-pack Rails work. API recall 27%.