Archived Charts
Historical benchmarks preserved for reference. These charts remain interactive, but their scores no longer contribute to current IQ or EQ.
Original ProofBench
Historical results combine benchmark versions and are excluded from IQ. Use the separately versioned ProofBench 1.1 for current comparisons.
Archived chartProofBench Benchmark ScoresArchived chart. Historical results combine benchmark versions and are excluded from IQ. Use the separately versioned ProofBench 1.1 for current comparisons.Data: Vals.aiOpen chartArchived chartProofBench Cost EfficiencyArchived chart. Historical results combine benchmark versions and are excluded from IQ. Use the separately versioned ProofBench 1.1 for current comparisons.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +1 moreOpen chartArchived chartProofBench vs Effective CostArchived chart. Historical results combine benchmark versions and are excluded from IQ. Use the separately versioned ProofBench 1.1 for current comparisons.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +1 moreOpen chart
Original Toolathlon
Historical results combine original and Verified cohorts and are excluded from IQ. Retained for reference; scores are not a consistent current-model comparison.
Archived chartToolathlon Benchmark ScoresArchived chart. Historical results combine original and Verified cohorts and are excluded from IQ. Retained for reference; scores are not a consistent current-model comparison.Data: LLM Stats, Toolathlon, LLM Stats ToolathlonOpen chartArchived chartToolathlon Cost EfficiencyArchived chart. Historical results combine original and Verified cohorts and are excluded from IQ. Retained for reference; scores are not a consistent current-model comparison.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +4 moreOpen chartArchived chartToolathlon vs Effective CostArchived chart. Historical results combine original and Verified cohorts and are excluded from IQ. Retained for reference; scores are not a consistent current-model comparison.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +4 moreOpen chart
Original AA-LCR
Superseded by AA-LCR v1.1 with corrected grading. These original scores are excluded from IQ and are not interchangeable with v1.1.
Archived chartAA Long Context Reasoning ScoresArchived chart. Superseded by AA-LCR v1.1 with corrected grading. These original scores are excluded from IQ and are not interchangeable with v1.1.Data: Artificial Analysis, Artificial Analysis model leaderboardOpen chartArchived chartAA Long Context Reasoning Cost EfficiencyArchived chart. Superseded by AA-LCR v1.1 with corrected grading. These original scores are excluded from IQ and are not interchangeable with v1.1.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chartArchived chartAA Long Context Reasoning vs Effective CostArchived chart. Superseded by AA-LCR v1.1 with corrected grading. These original scores are excluded from IQ and are not interchangeable with v1.1.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chart
FrontierSWE V1
V1 Dominance retired from Composite IQ and Software Engineering; V2 mean@5 requires a separate calibration.
Archived chartFrontierSWE ScoresArchived chart. V1 Dominance retired from Composite IQ and Software Engineering; V2 mean@5 requires a separate calibration.Data: FrontierSWEOpen chartArchived chartFrontierSWE Cost EfficiencyArchived chart. V1 Dominance retired from Composite IQ and Software Engineering; V2 mean@5 requires a separate calibration.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chartArchived chartFrontierSWE vs Effective CostArchived chart. V1 Dominance retired from Composite IQ and Software Engineering; V2 mean@5 requires a separate calibration.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chart
IOI V1
IOI V1 archived; V2 uses a different harness and 2024–2026 tasks.
Archived chartIOI Benchmark ScoresArchived chart. IOI V1 archived; V2 uses a different harness and 2024–2026 tasks.Data: Vals.ai IOIOpen chartArchived chartIOI Cost EfficiencyArchived chart. IOI V1 archived; V2 uses a different harness and 2024–2026 tasks.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chartArchived chartIOI vs Effective CostArchived chart. IOI V1 archived; V2 uses a different harness and 2024–2026 tasks.Data: Artificial Analysis, ARC Prize, Vals Index v2 Cost per Test +2 moreOpen chart