Skip to content

Benchmarks

Sources
8
Entries tracked
22,318
Last sync
24 min
All sources syncing normally
  • Arena · Text3 hours ago
  • Arena · WebDev3 hours ago
  • Arena · Vision3 hours ago
  • Arena · Agent3 hours ago
  • GPQA Diamond24 minutes ago
  • SWE-bench24 minutes ago
  • Terminal-Bench24 minutes ago
  • ARC-AGI-224 minutes ago
AIME (math) ×16 of 200 models
ModelOrg± CISwingAdd to compare
23claude-opus-4-8AnthropicSep 2026$5 / $2598.3%96.9%–99.7%#15–33
29claude-opus-4-7AnthropicSep 2026$5 / $2597.8%95.6%–100.0%#1–18
43claude-opus-4-6AnthropicSep 2026$5 / $2594.4%91.6%–97.3%#2–25
74claude-opus-4-5-20251101AnthropicNov 2025$5 / $2586.1%81.8%–90.4%#19–62
102claude-opus-4-1-20250805AnthropicAug 2025$15 / $7568.9%61.9%–75.9%#73–404
110claude-opus-4-20250514AnthropicMay 2025—64.4%57.2%–71.7%#84–470

Showing AIME (math) exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.

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.