GPQA Diamond — Epoch AI harness Effort: Low Configuration: Epoch AI harness 75.3% epoch.ai for OpenAI o4-mini (low) on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Evaluated July 11, 2026 at 8:04 AM UTC Benchmark leaderboard Epoch AI Independent Record for OpenAI o4-mini on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Epoch AI harness Effort: Medium Configuration: Epoch AI harness 77.8% epoch.ai for OpenAI o4-mini (medium) on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Evaluated August 7, 2026 at 12:24 AM UTC Benchmark leaderboard Epoch AI Independent Record for OpenAI o4-mini on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Epoch AI harness Effort: High Configuration: Epoch AI harness 79.6% epoch.ai for OpenAI o4-mini (high) on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Evaluated April 16, 2025 at 6:17 PM UTC Benchmark leaderboard Epoch AI Independent Record for OpenAI o4-mini on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Vals AI harness Effort: High Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 74.5% vals.ai for OpenAI o4-mini (high) on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for OpenAI o4-mini on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HealthBench Effort: High 50.1% deploymentsafety.openai.com for OpenAI o4-mini (high) on HealthBench HealthBench (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for OpenAI o4-mini on HealthBench HealthBench HealthBench HealthBench Consensus Effort: High 91.8% deploymentsafety.openai.com for OpenAI o4-mini (high) on HealthBench HealthBench Consensus (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for OpenAI o4-mini on HealthBench HealthBench Consensus HealthBench HealthBench Hard Effort: High 17.5% deploymentsafety.openai.com for OpenAI o4-mini (high) on HealthBench HealthBench Hard (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for OpenAI o4-mini on HealthBench HealthBench Hard Humanity's Last Exam Full set — Scale AI evaluation Effort: Medium 14.3% labs.scale.com for OpenAI o4-mini (medium) on Humanity's Last Exam Full set — Scale AI evaluation (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 3, 2026 Benchmark leaderboard Scale AI Independent Record for OpenAI o4-mini on Humanity's Last Exam Full set — Scale AI evaluation Humanity's Last Exam Full set — Scale AI evaluation Effort: High 18.1% labs.scale.com for OpenAI o4-mini (high) on Humanity's Last Exam Full set — Scale AI evaluation (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 3, 2026 Benchmark leaderboard Scale AI Independent Record for OpenAI o4-mini on Humanity's Last Exam Full set — Scale AI evaluation Humanity's Last Exam Text only — Scale AI evaluation Effort: Medium 14.5% labs.scale.com for OpenAI o4-mini (medium) on Humanity's Last Exam Text only — Scale AI evaluation (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for OpenAI o4-mini on Humanity's Last Exam Text only — Scale AI evaluation Humanity's Last Exam Text only — Scale AI evaluation Effort: High 18.9% labs.scale.com for OpenAI o4-mini (high) on Humanity's Last Exam Text only — Scale AI evaluation (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for OpenAI o4-mini on Humanity's Last Exam Text only — Scale AI evaluation MMLU-Pro Original — Vals AI harness Effort: High Configuration: Vals AI harness 80.6% vals.ai for OpenAI o4-mini (high) on MMLU-Pro Original — Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for OpenAI o4-mini on MMLU-Pro Original — Vals AI harness MMMU-Pro Vals AI harness Effort: High Configuration: Vals AI harness 79.7% vals.ai for OpenAI o4-mini (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for OpenAI o4-mini on MMMU-Pro Vals AI harness SimpleQA Verified Original — Epoch AI harness Effort: Low Configuration: Epoch AI harness 19.6% epoch.ai for OpenAI o4-mini (low) on SimpleQA Verified Original — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported October 7, 2026 Evaluated August 27, 2026 at 7:29 PM UTC Benchmark leaderboard Epoch AI Independent Record for OpenAI o4-mini on SimpleQA Verified Original — Epoch AI harness SimpleQA Verified Original — Epoch AI harness Effort: High Configuration: Epoch AI harness 18.8% epoch.ai for OpenAI o4-mini (high) on SimpleQA Verified Original — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported October 7, 2026 Evaluated August 27, 2026 at 7:28 PM UTC Benchmark leaderboard Epoch AI Independent Record for OpenAI o4-mini on SimpleQA Verified Original — Epoch AI harness SWE-bench Verified — mini-swe-agent Configuration: mini-swe-agent harness 45.0% github.com for OpenAI o4-mini on SWE-bench Verified — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported July 26, 2025 Benchmark repository SWE-bench Independent SWE-bench Team Record for OpenAI o4-mini on SWE-bench Verified — mini-swe-agent Toolathlon Original 14.8% toolathlon.xyz for OpenAI o4-mini on Toolathlon Original (opens in a new tab) Archived copy (opens in a new tab) Tasks completed Reported October 28, 2025 Benchmark leaderboard Toolathlon Independent Toolathlon Authors Record for OpenAI o4-mini on Toolathlon Original