| GPQA diamondKnowledge | 95.39%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · gpqa_diamond.csv Reported standard error: ±1.40 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| FrontierMath-Tiers-1-3-v2-PrivateReasoning | 68.42%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · frontiermath_tiers_1_3_v2.csv Reported standard error: ±2.76 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| FrontierMath-Tier-4-v2-PrivateReasoning | 21.95%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · frontiermath_tier_4_v2.csv Reported standard error: ±6.54 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| OTIS Mock AIME 2024-2025Reasoning | 98.89%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · otis_mock_aime_2024_2025.csv Reported standard error: ±0.87 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| SimpleQA VerifiedKnowledge | 69.70%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · simpleqa_verified.csv Reported standard error: ±1.45 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| Chess PuzzlesReasoning | 58.00%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-10-13 · chess_puzzles.csv Reported standard error: ±4.96 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| Mystery Game PuzzlesReasoning | 47.00%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · mystery_game_puzzles.csv Reported standard error: ±5.02 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| ARC-AGIReasoning | 98.50%gemini-3.8-flash_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 | 97.50%gemini-3.8-flash_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 | 90.50%gemini-3.8-flash_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 ↗ |
|---|
| WeirdMLCoding | 84.81%gemini-3.8-flash_high https://htihle.github.io/weirdml.html · external Run: Date not supplied · weirdml_external.csv Reported standard error: ±0.01 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| APEX-AgentsAgents | 64.30%gemini-3.8-flash_unknown External result; evaluator not specified · external Run: Date not supplied · apex_agents_external.csv Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| ARC-AGI-2Reasoning | 89.17%gemini-3.8-flash_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 | 82.92%gemini-3.8-flash_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 | 77.50%gemini-3.8-flash_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 ↗ |
|---|
| HLEKnowledge | 44.52%gemini-3.8-flash_unknown External result; evaluator not specified · external Run: Date not supplied · hle_external.csv Reported standard error: ±1.96 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| CursorBenchOther | 39.60%gemini-3.8-flash_high https://cursor.com/cursorbench · external Run: Date not supplied · cursorbench_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| CursorBenchOther | 37.30%gemini-3.8-flash_medium https://cursor.com/cursorbench · external Run: Date not supplied · cursorbench_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| CritPtOther | 18.29%gemini-3.8-flash_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 | 12.29%gemini-3.8-flash_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 | 4.00%gemini-3.8-flash_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 ↗ |
|---|
| Remote Labor IndexAgents | 5.83%gemini-3.8-flash_unknown External result; evaluator not specified · external Run: Date not supplied · rli_external.csv Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| Surface Evolver BenchOther | 76.88%gemini-3.8-flash_high https://yhenon.github.io/surface-evolver-llm-eval/ · external Run: Date not supplied · surface_evolver_bench_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| DeepSWECoding | 73.83%gemini-3.8-flash_high https://deepswe.datacurve.ai/ · external Run: Date not supplied · deepswe_external.csv Inspect source ↗CC BY 4.0 | high / mini-swe-agent | External · via Epoch ↗ |
|---|
| DeepSWECoding | 71.02%gemini-3.8-flash_medium https://deepswe.datacurve.ai/ · external Run: Date not supplied · deepswe_external.csv Inspect source ↗CC BY 4.0 | medium / mini-swe-agent | External · via Epoch ↗ |
|---|
| ProofBenchReasoning | 48.00%gemini-3.8-flash_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 | 41.19%gemini-3.8-flash_medium https://cognition.com/frontiercode · external Run: Date not supplied · frontiercode_external.csv Inspect source ↗CC BY 4.0 | medium / devin-cli | External · via Epoch ↗ |
|---|
| LMCAReasoning | 52.90%gemini-3.8-flash_medium https://conceptualreasoning.ai/lmca · external Run: Date not supplied · lmca_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| LMCAReasoning | 51.55%gemini-3.8-flash_unknown https://conceptualreasoning.ai/lmca · external Run: Date not supplied · lmca_external.csv Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| LMCAReasoning | 51.16%gemini-3.8-flash_low https://conceptualreasoning.ai/lmca · external Run: Date not supplied · lmca_external.csv Inspect source ↗CC BY 4.0 | low / Not reported | External · via Epoch ↗ |
|---|
| DTBenchReasoning | 94.93%gemini-3.8-flash_medium https://conceptualreasoning.ai/dtbench · external Run: Date not supplied · dtbench_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| DTBenchReasoning | 95.72%gemini-3.8-flash_unknown https://conceptualreasoning.ai/dtbench · external Run: Date not supplied · dtbench_external.csv Inspect source ↗CC BY 4.0 | unknown / Not reported | External · via Epoch ↗ |
|---|
| DTBenchReasoning | 94.93%gemini-3.8-flash_low https://conceptualreasoning.ai/dtbench · external Run: Date not supplied · dtbench_external.csv Inspect source ↗CC BY 4.0 | low / Not reported | External · via Epoch ↗ |
|---|
| FrontierMath-ErdosReasoning | 0.00%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-02 · frontiermath_erdos.csv Reported standard error: ±0.00 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| Furniture AssemblyReasoning | 31.67%gemini-3.8-flash_high Epoch AI evaluation · epoch-run Run: 2026-09-10 · furniture_assembly.csv Reported standard error: ±5.94 pp. This is not a 90% interval. Inspect source ↗CC BY 4.0 | high / Epoch Inspect | Epoch AI run ↗ |
|---|
| GDP.pdfOther | 23.40%gemini-3.8-flash_medium https://surgehq.ai/benchmarks/gdp-pdf · external Run: Date not supplied · gdp_pdf_external.csv Inspect source ↗CC BY 4.0 | medium / Not reported | External · via Epoch ↗ |
|---|
| GDP.pdfOther | 23.20%gemini-3.8-flash_high https://surgehq.ai/benchmarks/gdp-pdf · external Run: Date not supplied · gdp_pdf_external.csv Inspect source ↗CC BY 4.0 | high / Not reported | External · via Epoch ↗ |
|---|
| FrontierSWECoding | 19.63%gemini-3.8-flash_high https://www.frontierswe.com/ · external Run: Date not supplied · frontierswe_external.csv Inspect source ↗CC BY 4.0 | high / proximus | External · via Epoch ↗ |
|---|
| WeirdML v3Coding | 14.57%gemini-3.8-flash_high https://htihle.github.io/weirdml.html · external Run: Date not supplied · weirdml_v3_external.csv Inspect source ↗CC BY 4.0 | high / gemini_cli | External · via Epoch ↗ |
|---|
| LiveBench 2 Coding · 20261025Coding | 82.20%gemini-3.8-flash · high · LiveBench 2 20261025 LiveBench · independent-run Run: Date not supplied · 20261025/table.csv Benchmark edition 25 October 2026; individual run date not supplied. One run per model; mean task score. Failed/unmeasurable tasks score zero in the source. Source effort metadata is preserved. Inspect source ↗Attributed public leaderboard score facts · LiveBench | high / LiveBench 2 · 20261025 | LiveBench ↗ |
|---|
| LiveBench 2 Reasoning · 20261025Reasoning | 89.30%gemini-3.8-flash · high · LiveBench 2 20261025 LiveBench · independent-run Run: Date not supplied · 20261025/table.csv Benchmark edition 25 October 2026; individual run date not supplied. One run per model; mean task score. Failed/unmeasurable tasks score zero in the source. Source effort metadata is preserved. Inspect source ↗Attributed public leaderboard score facts · LiveBench | high / LiveBench 2 · 20261025 | LiveBench ↗ |
|---|
| LiveBench 2 Long Agentic · 20261025Agents | 16.50%gemini-3.8-flash · high · LiveBench 2 20261025 LiveBench · independent-run Run: Date not supplied · 20261025/table.csv Benchmark edition 25 October 2026; individual run date not supplied. One run per model; mean task score. Failed/unmeasurable tasks score zero in the source. Source effort metadata is preserved. Inspect source ↗Attributed public leaderboard score facts · LiveBench | high / LiveBench 2 · 20261025 | LiveBench ↗ |
|---|
| Terminal-Bench 4.0Agents | 19.09%Gemini 3.8 Flash · mini-SWE-agent · high Terminal-Bench public leaderboard · benchmark-publisher Run: 2026-09-02 · 4-0-0/06850434-507d-4dbd-b74a-09d52519ee35 Published model + agent result; agents differ across models. Not a controlled model-only evaluation. Reported trials: 330. Source 95% accuracy interval half-width: 3.36 pp; this is not a DC Power interval. Inspect source ↗Attributed public leaderboard score facts · Terminal-Bench | high / Terminal-Bench 4.0 · mini-SWE-agent | Terminal-Bench ↗ |
|---|
| ARC-AGI-2 · semi-privateReasoning | 89.17%Gemini 3.8 Flash (High) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.399729 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 | 82.92%Gemini 3.8 Flash (Medium) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.247788 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 | 77.50%Gemini 3.8 Flash (Low) ARC Prize official leaderboard · benchmark-publisher Run: Date not supplied · evaluations.json · v2_Semi_Private Reported benchmark cost: $0.154454 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 ↗ |
|---|