Gemini 3.5 Flash beats 3.1 Pro on the benchmarks that matter

The other I/O headline: Gemini 3.5 Flash — Google's new Flash-tier model — outperformed Gemini 3.1 Pro on three coding and agentic benchmarks Google highlighted: Terminal-Bench 2.1 (76.2%), GDPval-AA (1,656 Elo), and MCP Atlas (83.6%). Google also reported output throughput roughly 4× that of other frontier models. At launch, 3.5 Flash was generally available through Antigravity, the Gemini API, Google AI Studio, and Android Studio.

Reception is mixed-positive. The long-standing "Gemini feels lazy" complaint has reportedly mostly faded in early testing, sub-200ms responses on many prompts make it feel genuinely real-time, and LM Arena coding scores have it ahead of 3.1 Pro at meaningfully lower per-token cost. The honest counterweight: Hacker News threads on the prior 3.x releases still surface a steady drumbeat of "Gemini is consistently the most frustrating model I use" from developers — benchmarks aren't the same as daily-driver feel, and Google hasn't fully closed that gap yet.

Practical read: two things matter here. First, "Flash beats Pro" inverts the normal model hierarchy on the benchmarks Google selected. The cheap-and-fast tier is no longer automatically a worse version of the expensive one; it is a different point on the price/capability curve that sometimes wins. Second, GDPval-AA and MCP Atlas are agent-shaped benchmarks — tool use and long-horizon tasks. A Flash-tier model leading there lowers the expected cost floor for capable agent runtimes, although real workloads still need their own evaluation.