AI Benchmark Digest — 2026-07-05

View on AI Benchmark Hub

Daily

New #1 Leaders (3)

• Surface Evolver Bench Pass Rate (Pass Rate (%)): Claude Fable 5 (High) (87.5) beat GPT-5.5 (High) (75.0) by 12.5.
• Surface Evolver Bench (Mean Score (%)): Claude Fable 5 (High) (95.0) beat GPT-5.5 (High) (88.12) by 6.88.
• FutureEval (Unified Forecasting Score): Gemini 3.1 Pro (Preview) (High) (19.84) beat Claude Opus 4.5 High 32k (19.54) by 0.3.