APEX-Agents Original Effort: High 4.4% mercor.com for gpt-oss-120b (high) on APEX-Agents Original (opens in a new tab) Archived copy (opens in a new tab) Pass rate Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for gpt-oss-120b on APEX-Agents Original AutomationBench 1.0.6 Effort: High 0.2% zapier.com for gpt-oss-120b (high) on AutomationBench 1.0.6 (opens in a new tab) Archived copy (opens in a new tab) Task pass rate Reported October 7, 2026 Benchmark leaderboard Zapier Independent Zapier AutomationBench Team Record for gpt-oss-120b on AutomationBench 1.0.6 Codeforces No tools Effort: Low Configuration: No tools 1595 deploymentsafety.openai.com for gpt-oss-120b (low) on Codeforces No tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces No tools Codeforces No tools Effort: Medium Configuration: No tools 2205 deploymentsafety.openai.com for gpt-oss-120b (medium) on Codeforces No tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces No tools Codeforces No tools Effort: High Configuration: No tools 2463 deploymentsafety.openai.com for gpt-oss-120b (high) on Codeforces No tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces No tools Codeforces With tools Effort: Low Configuration: With tools 1653 deploymentsafety.openai.com for gpt-oss-120b (low) on Codeforces With tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces With tools Codeforces With tools Effort: Medium Configuration: With tools 2365 deploymentsafety.openai.com for gpt-oss-120b (medium) on Codeforces With tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces With tools Codeforces With tools Effort: High Configuration: With tools 2622 deploymentsafety.openai.com for gpt-oss-120b (high) on Codeforces With tools (opens in a new tab) Codeforces Elo Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Codeforces With tools DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Effort: High Configuration: Mercor mini-swe-agent, 500 steps, 2 hours 0.0% mercor.com for gpt-oss-120b (high) on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps (opens in a new tab) Archived copy (opens in a new tab) DeepSWE Pass@1 Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for gpt-oss-120b on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps GPQA Diamond Effort: Low Reported as: low (no tools) Configuration: No tools 67.1% deploymentsafety.openai.com for gpt-oss-120b (low (no tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond Effort: Low Reported as: low (with tools) Configuration: With tools 68.1% deploymentsafety.openai.com for gpt-oss-120b (low (with tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond Effort: Medium Reported as: medium (no tools) Configuration: No tools 73.1% deploymentsafety.openai.com for gpt-oss-120b (medium (no tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond Effort: Medium Reported as: medium (with tools) Configuration: With tools 73.5% deploymentsafety.openai.com for gpt-oss-120b (medium (with tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond Effort: High Reported as: high (no tools) Configuration: No tools 80.1% deploymentsafety.openai.com for gpt-oss-120b (high (no tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond Effort: High Reported as: high (with tools) Configuration: With tools 80.9% deploymentsafety.openai.com for gpt-oss-120b (high (with tools)) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond — Epoch AI harness Effort: High Configuration: Epoch AI harness 75.8% epoch.ai for gpt-oss-120b (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 December 11, 2025 at 10:37 PM UTC Benchmark leaderboard Epoch AI Independent Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 77.5% mercor.com for gpt-oss-120b (high) on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for gpt-oss-120b on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot GPQA Diamond — Vals AI harness Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 78.5% vals.ai for gpt-oss-120b 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-oss-120b on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HealthBench Effort: Low 53.0% deploymentsafety.openai.com for gpt-oss-120b (low) on HealthBench HealthBench (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench HealthBench Effort: Medium 55.9% deploymentsafety.openai.com for gpt-oss-120b (medium) on HealthBench HealthBench (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench HealthBench Effort: High 57.6% deploymentsafety.openai.com for gpt-oss-120b (high) on HealthBench HealthBench (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench HealthBench HealthBench Consensus Effort: Low 90.6% deploymentsafety.openai.com for gpt-oss-120b (low) on HealthBench HealthBench Consensus (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Consensus HealthBench HealthBench Consensus Effort: Medium 90.8% deploymentsafety.openai.com for gpt-oss-120b (medium) on HealthBench HealthBench Consensus (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Consensus HealthBench HealthBench Consensus Effort: High 89.9% deploymentsafety.openai.com for gpt-oss-120b (high) on HealthBench HealthBench Consensus (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Consensus HealthBench HealthBench Hard Effort: Low 22.8% deploymentsafety.openai.com for gpt-oss-120b (low) on HealthBench HealthBench Hard (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Hard HealthBench HealthBench Hard Effort: Medium 26.9% deploymentsafety.openai.com for gpt-oss-120b (medium) on HealthBench HealthBench Hard (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Hard HealthBench HealthBench Hard Effort: High 30.0% deploymentsafety.openai.com for gpt-oss-120b (high) on HealthBench HealthBench Hard (opens in a new tab) HealthBench Score Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on HealthBench HealthBench Hard Humanity's Last Exam Text only — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 18.5% mercor.com for gpt-oss-120b (high) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for gpt-oss-120b on Humanity's Last Exam Text only — Mercor single-shot Humanity's Last Exam Text only — Scale AI evaluation 15.5% labs.scale.com for gpt-oss-120b 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-oss-120b on Humanity's Last Exam Text only — Scale AI evaluation MMLU Effort: Low 85.9% deploymentsafety.openai.com for gpt-oss-120b (low) on Massive Multitask Language Understanding MMLU (opens in a new tab) MMLU Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Massive Multitask Language Understanding MMLU MMLU Effort: Medium 88.0% deploymentsafety.openai.com for gpt-oss-120b (medium) on Massive Multitask Language Understanding MMLU (opens in a new tab) MMLU Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Massive Multitask Language Understanding MMLU MMLU Effort: High 90.0% deploymentsafety.openai.com for gpt-oss-120b (high) on Massive Multitask Language Understanding MMLU (opens in a new tab) MMLU Accuracy Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on Massive Multitask Language Understanding MMLU MMLU-Pro Original — Vals AI harness Configuration: Vals AI harness 79.2% vals.ai for gpt-oss-120b 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-oss-120b on MMLU-Pro Original — Vals AI harness SWE-bench Verified Effort: Low 47.9% deploymentsafety.openai.com for gpt-oss-120b (low) on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on SWE-bench Verified SWE-bench Verified Effort: Medium 52.6% deploymentsafety.openai.com for gpt-oss-120b (medium) on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on SWE-bench Verified SWE-bench Verified Effort: High 62.4% deploymentsafety.openai.com for gpt-oss-120b (high) on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported August 5, 2025 Model card OpenAI Self-reported Record for gpt-oss-120b on SWE-bench Verified SWE-bench Verified — mini-swe-agent Configuration: mini-swe-agent harness 26.0% github.com for gpt-oss-120b on SWE-bench Verified — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported August 7, 2025 Benchmark repository SWE-bench Independent SWE-bench Team Record for gpt-oss-120b on SWE-bench Verified — mini-swe-agent SWE-bench Verified — Vals AI mini-swe-agent Configuration: mini-swe-agent harness 33.6% vals.ai for gpt-oss-120b on SWE-bench Verified — Vals AI mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for gpt-oss-120b on SWE-bench Verified — Vals AI mini-swe-agent SWE-bench Pro Public — Scale AI SWE-Agent Configuration: SWE-Agent harness, 250 turns, uncapped cost 16.2% labs.scale.com for gpt-oss-120b on SWE-bench Pro Public — Scale AI SWE-Agent (opens in a new tab) Archived copy (opens in a new tab) Public tasks resolved Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for gpt-oss-120b on SWE-bench Pro Public — Scale AI SWE-Agent Terminal-Bench 2.0 — Vals AI Terminus 2 Configuration: Terminus-2 harness 19.1% vals.ai for gpt-oss-120b 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-oss-120b on Terminal-Bench 2.0 — Vals AI Terminus 2