| GPQA diamondKnowledge | 94.32%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · gpqa_diamond.csv Reported standard error: ±1.49 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| FrontierMath-Tiers-1-3-v2-PrivateReasoning | 89.82%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · frontiermath_tiers_1_3_v2.csv Reported standard error: ±1.79 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| FrontierMath-Tier-4-v2-PrivateReasoning | 90.00%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · frontiermath_tier_4_v2.csv Reported standard error: ±5.17 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| OTIS Mock AIME 2024-2025Reasoning | 100.00%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · otis_mock_aime_2024_2025.csv Reported standard error: ±0.00 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| SimpleQA VerifiedKnowledge | 60.70%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · simpleqa_verified.csv Reported standard error: ±1.55 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| Chess PuzzlesReasoning | 56.00%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-10-13 · chess_puzzles.csv Reported standard error: ±4.99 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| EBR-benchOther | 53.33%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · ebr_bench.csv Reported standard error: ±4.62 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| Mystery Game PuzzlesReasoning | 56.00%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · mystery_game_puzzles.csv Reported standard error: ±4.99 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| ARC-AGIReasoning | 95.50%gpt-6-sol_max https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGIReasoning | 92.67%gpt-6-sol_xhigh https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | xhigh / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGIReasoning | 91.00%gpt-6-sol_high https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGIReasoning | 83.67%gpt-6-sol_medium https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGIReasoning | 72.17%gpt-6-sol_low https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | low / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGIReasoning | 29.33%gpt-6-sol_none https://arcprize.org/leaderboard · external Run: Date not supplied · arc_agi_external.csv Inspect source ↗CC BY 4.0 | none / Not reported | External · via Epoch ↗ |
|---|
| APEX-AgentsAgents | 54.30%gpt-6-sol_max External result; evaluator not specified · external Run: Date not supplied · apex_agents_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 89.58%gpt-6-sol_max External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 78.06%gpt-6-sol_xhigh External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | xhigh / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 68.89%gpt-6-sol_high External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 57.78%gpt-6-sol_medium External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 31.53%gpt-6-sol_low External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | low / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 1.67%gpt-6-sol_none External result; evaluator not specified · external Run: Date not supplied · arc_agi_2_external.csv Inspect source ↗CC BY 4.0 | none / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 30.86%gpt-6-sol_max External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 28.00%gpt-6-sol_xhigh External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | xhigh / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 25.43%gpt-6-sol_high External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 24.57%gpt-6-sol_medium External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 16.29%gpt-6-sol_low External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | low / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 4.00%gpt-6-sol_none External result; evaluator not specified · external Run: Date not supplied · critpt_external.csv Inspect source ↗CC BY 4.0 | none / Not reported | External · via Epoch ↗ |
|---|
| ProofBenchReasoning | 83.00%gpt-6-sol_unknown https://www.vals.ai/benchmarks/proof_bench · external Run: Date not supplied · proofbench_external.csv Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| FrontierCodeCoding | 49.27%gpt-6-sol_max https://cognition.com/frontiercode · external Run: Date not supplied · frontiercode_external.csv Inspect source ↗CC BY 4.0 | max / codex | External · via Epoch ↗ |
|---|
| LMCAReasoning | 59.08%gpt-6-sol_max https://conceptualreasoning.ai/lmca · external Run: Date not supplied · lmca_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| DTBenchReasoning | 97.33%gpt-6-sol_max https://conceptualreasoning.ai/dtbench · external Run: Date not supplied · dtbench_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| FrontierMath-ErdosReasoning | 1.47%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-30 · frontiermath_erdos.csv Reported standard error: ±1.47 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| Furniture AssemblyReasoning | 58.33%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-23 · furniture_assembly.csv Reported standard error: ±6.42 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| GDP.pdfOther | 26.40%gpt-6-sol_max https://surgehq.ai/benchmarks/gdp-pdf · external Run: Date not supplied · gdp_pdf_external.csv Inspect source ↗CC BY 4.0 | max / Not reported | External · via Epoch ↗ |
|---|
| WeirdML v3Coding | 27.56%gpt-6-sol_xhigh https://htihle.github.io/weirdml.html · external Run: Date not supplied · weirdml_v3_external.csv Inspect source ↗CC BY 4.0 | xhigh / codex_cli | External · via Epoch ↗ |
|---|
| SpireClimb (Slay the Spire 1, Ascension 20 Heart)Other | 5.00%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-11-05 · spireclimb_sts1_a20_heart.csv Reported standard error: ±2.84 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| SpireClimb (Slay the Spire 2, Ascension 10)Other | 13.33%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-09-22 · spireclimb_sts2_a10.csv Reported standard error: ±3.95 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| SpireClimbOther | 9.17%gpt-6-sol_max Epoch AI evaluation · epoch-run Run: 2026-11-05 · spireclimb.csv Reported standard error: ±2.43 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | max / Epoch Inspect | Epoch AI run ↗ |
|---|
| Terminal-Bench 4.0Agents | 49.39%GPT-6 Sol · Codex · max Terminal-Bench public leaderboard · benchmark-publisher Run: 2026-09-22 · 4-0-0/a29d2361-9d5b-4a95-a56d-707070bd6c88 Published model + agent result; agents differ across models. Not a controlled model-only evaluation. Reported trials: 330. Source 95% accuracy interval half-width: 3.23 pp; this is not a DC Power interval. Inspect source ↗Attributed public leaderboard score facts · Terminal-Bench | max / Terminal-Bench 4.0 · Codex | Terminal-Bench ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 89.58%GPT-6 Sol (Max) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.438898 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | max / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 78.06%GPT-6 Sol (XHigh) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.277229 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | xhigh / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 68.89%GPT-6 Sol (High) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.202737 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | high / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 57.78%GPT-6 Sol (Medium) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.147627 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | medium / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 31.53%GPT-6 Sol (Low) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.100858 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | low / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 1.67%GPT-6 Sol (None) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.074064 per task. Different from the token-workload estimate. Semi-private split only. Provider-adapter and custom-scaffold families are kept separate. Identical effort names do not imply identical compute budgets. Inspect source ↗Public leaderboard facts · ARC Prize | none / ARC Prize leaderboard baseline | ARC Prize ↗ |
|---|