Agent Arena Cost Efficiency
Source-matched Agent Arena Net Improvement and median session-cost configurations from the methodology window. Each line is one canonical model. Color = provider.
Agent Arena Cost Efficiency
Agent Arena Cost Efficiency
Source-matched Agent Arena Net Improvement and median session-cost configurations from the methodology window. Each line is one canonical model. Color = provider.
How to read this chart
Each line uses the shared model-configuration dataset to compare source-matched reasoning-effort levels on Agent Arena. Every point shows the score and median cost per session for that exact configuration. Canonical score bars and bell curves remain one point per model.
Data sources
More / ExperimentalAgent Arena vs Effective CostX = effective cost (log). Y = Agent Arena Net Improvement. Color = provider.Data: Artificial Analysis, ARC Prize, Vals.ai +1 moreOpen chartIQ BenchmarksArena.ai Agent Arena ScoresAggregate net improvement across real-world Agent Mode sessions. Color = provider.Data: Arena.ai Agent ArenaOpen chartMore / ExperimentalTerminal-Bench 2.0 Benchmark ScoresLegacy Terminal-Bench 2.0 scores retained for historical comparison. Color = provider.Data: Terminal-BenchOpen chartMore / ExperimentalARC-AGI-3 Cost EfficiencyX = ARC Prize reported Cost (V3) (log). Y = ARC-AGI-3 %. Each line connects one model's published reasoning-effort levels, so the score-vs-cost tradeoff is visible per model. Defaults to the current model generation. Color = provider.Data: ARC PrizeOpen chartMore / ExperimentalARC-AGI-2 Cost EfficiencyX = ARC Prize reported cost/task (log). Y = ARC-AGI-2 %. Each line connects one model's published reasoning-effort levels, so the score-vs-cost tradeoff is visible per model. Defaults to the current model generation. Color = provider.Data: ARC PrizeOpen chartMore / ExperimentalARC-AGI-1 Cost EfficiencyX = ARC Prize reported cost/task (log). Y = ARC-AGI-1 %. Each line connects one model's published reasoning-effort levels, so the score-vs-cost tradeoff is visible per model. Defaults to the current model generation. Color = provider.Data: ARC PrizeOpen chart