Skip to content

Benchmarks

Sources
8
Entries tracked
22,318
Last sync
27 min
All sources syncing normally
  • Arena · Text48 minutes ago
  • Arena · WebDev48 minutes ago
  • Arena · Vision48 minutes ago
  • Arena · Agent48 minutes ago
  • GPQA Diamond27 minutes ago
  • SWE-bench27 minutes ago
  • Terminal-Bench27 minutes ago
  • ARC-AGI-227 minutes ago
Overall ×1WebDev Arena ×1Vision Arena ×1Factuality Arena ×1Search Arena ×1Document Arena ×1Agent Arena ×1GPQA Diamond ×1AIME (math) ×1FrontierMath ×1SimpleQA Verified ×1ARC-AGI-2 ×1Humanity's Last Exam ×1SimpleBench ×1LiveBench ×1BFCL v4 (function calling) ×1dormantSWE-bench Verified (live board) ×1Terminal-Bench 4.0 ×1MMLU-Pro (Artificial Analysis) ×1LiveCodeBench (Artificial Analysis) ×1IFBench (Artificial Analysis) ×1HHEM Hallucination (Vectara) ×12 of 1034 models
ModelOrgIntervalSwingiiiiiiiiiiiiiiiiiiiiiiAdd to compare
371gemini-2.5-flash-preview-09-2025-highthinkingGoogle——51.6 56.4Ranked 51.6 = (1 × 56.4 + 3 × 50) ÷ 4. 56.4 is the mean percentile over the 1 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.——1/22low———————————————56———————
383gemini-2.5-flash-preview-09-2025googleSep 2025—51.1 52.3Ranked 51.1 = (3 × 52.3 + 3 × 50) ÷ 6. 52.3 is the mean percentile over the 3 of 22 selected axes this model is measured on; fewer axes means a stronger pull toward 50.49–53293/22#141–38364—5835——————————————————

Blend is coverage-shrunk; the raw composite is shown greyed beside it · A low badge means fewer than 3 boards scored this model

The few things in AI and quantum that actually mattered this week — honestly summarised.

One email a week. Unsubscribe in one click. No tracking, no sharing.