Gemini 2.5 Flash

Gemini 2.5 Flash is Google's fast, cost-efficient multimodal large language model optimized for high-volume, low-latency workloads including agentic pipelines, coding, and general reasoning. It is Google's first Flash-series model to support optional 'thinking' (hybrid reasoning) that can be toggled or budget-capped by developers. Source: supabase.