GPQA Diamond 66.3% openai.com for GPT-4.1 on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported April 14, 2025 Launch post OpenAI Self-reported Record for GPT-4.1 on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond — Epoch AI harness Configuration: Epoch AI harness 66.9% epoch.ai for GPT-4.1 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 14, 2025 at 6:34 PM UTC Benchmark leaderboard Epoch AI Independent Record for GPT-4.1 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 65.4% vals.ai for GPT-4.1 (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 GPT-4.1 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness Humanity's Last Exam Full set — Scale AI evaluation 5.4% labs.scale.com for GPT-4.1 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 GPT-4.1 on Humanity's Last Exam Full set — Scale AI evaluation Humanity's Last Exam Text only — Scale AI evaluation 5.0% labs.scale.com for GPT-4.1 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 GPT-4.1 on Humanity's Last Exam Text only — Scale AI evaluation MMMU 74.8% openai.com for GPT-4.1 on Massive Multi-discipline Multimodal Understanding MMMU (opens in a new tab) MMMU Accuracy Reported April 14, 2025 Launch post OpenAI Self-reported Record for GPT-4.1 on Massive Multi-discipline Multimodal Understanding MMMU MMLU 90.2% openai.com for GPT-4.1 on Massive Multitask Language Understanding MMLU (opens in a new tab) MMLU Accuracy Reported April 14, 2025 Launch post OpenAI Self-reported Record for GPT-4.1 on Massive Multitask Language Understanding MMLU MMLU-Pro Original — Vals AI harness Effort: High Configuration: Vals AI harness 80.5% vals.ai for GPT-4.1 (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 GPT-4.1 on MMLU-Pro Original — Vals AI harness MMMU-Pro Vals AI harness Effort: High Configuration: Vals AI harness 72.4% vals.ai for GPT-4.1 (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 GPT-4.1 on MMMU-Pro Vals AI harness SimpleQA Verified Original — Epoch AI harness Configuration: Epoch AI harness 31.1% epoch.ai for GPT-4.1 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 31, 2026 at 10:01 PM UTC Benchmark leaderboard Epoch AI Independent Record for GPT-4.1 on SimpleQA Verified Original — Epoch AI harness SWE-bench Verified 54.6% openai.com for GPT-4.1 on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported April 14, 2025 Launch post OpenAI Self-reported Record for GPT-4.1 on SWE-bench Verified SWE-bench Verified — Epoch AI harness Configuration: Epoch AI harness 48.5% epoch.ai for GPT-4.1 on SWE-bench Verified — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported October 7, 2026 Evaluated February 8, 2026 at 5:41 PM UTC Benchmark leaderboard Epoch AI Independent Record for GPT-4.1 on SWE-bench Verified — Epoch AI harness SWE-bench Verified — mini-swe-agent Configuration: mini-swe-agent harness 39.6% github.com for GPT-4.1 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 GPT-4.1 on SWE-bench Verified — mini-swe-agent Terminal-Bench 2.0 — Vals AI Terminus 2 Effort: High Configuration: Terminus-2 harness 14.6% vals.ai for GPT-4.1 (high) on Terminal-Bench 2.0 — Vals AI Terminus 2 (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported June 4, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-4.1 on Terminal-Bench 2.0 — Vals AI Terminus 2