Artificial intelligence · May 2026
Gemini 3.5
A Flash model overtakes the previous generation's Pro.
Gemini 3.5 Flash beat Gemini 3.1 Pro on Terminal-Bench 2.1 (76.2%) and MCP Atlas (83.6%), showing how quickly frontier capability moves into cheap, fast tiers.
What it built on 1
Related measurements
The dashed line marks when it first worked.
Best AI score on GPQA Diamond Latest: 95.8 % correct (2026) Source: Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv); Anthropic: Claude 3 public launch (March 4, 2024)
Training compute of notable AI models Source: Epoch AI, Notable AI Models (CC BY 4.0)