Benchmarks

10 results across 7 benchmarks

Benchmark results for GPT-5.3-Codex
BenchmarkScoreSource
GDPvalEffort: xHigh70.9%openai.com for GPT-5.3-Codex (xhigh) on GDPval GDPval (opens in a new tab)Wins or tiesReported Launch postOpenAISelf-reportedRecord for GPT-5.3-Codex on GDPval GDPval
OSWorld VerifiedEffort: xHigh64.7%openai.com for GPT-5.3-Codex (xhigh) on OSWorld Verified (opens in a new tab)OSWorld-Verified success rateReported Launch postOpenAISelf-reportedRecord for GPT-5.3-Codex on OSWorld Verified
SWE Atlas Codebase QnA — Public; CodexEffort: xHighConfiguration: Codex harness32.6%labs.scale.com for GPT-5.3-Codex (xHigh) on SWE Atlas Codebase QnA — Public; Codex (opens in a new tab)Archived copy (opens in a new tab)Mean pass@1Reported Benchmark leaderboardScale AIIndependentRecord for GPT-5.3-Codex on SWE Atlas Codebase QnA — Public; Codex
SWE Atlas Refactoring — CodexEffort: xHighConfiguration: Codex harness42.4%labs.scale.com for GPT-5.3-Codex (xHigh) on SWE Atlas Refactoring — Codex (opens in a new tab)Archived copy (opens in a new tab)Mean pass@1Reported Benchmark leaderboardScale AIIndependentRecord for GPT-5.3-Codex on SWE Atlas Refactoring — Codex
SWE-bench Verified — Epoch AI harnessEffort: HighConfiguration: Epoch AI harness74.8%epoch.ai for GPT-5.3-Codex (high) on SWE-bench Verified — Epoch AI harness (opens in a new tab)Archived copy (opens in a new tab)Resolved TasksReported Evaluated Benchmark leaderboardEpoch AIIndependentRecord for GPT-5.3-Codex on SWE-bench Verified — Epoch AI harness
SWE-bench Verified — Vals AI mini-swe-agentEffort: xHighConfiguration: mini-swe-agent harness78.0%vals.ai for GPT-5.3-Codex (xhigh) on SWE-bench Verified — Vals AI mini-swe-agent (opens in a new tab)Archived copy (opens in a new tab)Resolved TasksReported Benchmark leaderboardVals AIIndependentRecord for GPT-5.3-Codex on SWE-bench Verified — Vals AI mini-swe-agent
SWE-bench Pro PublicEffort: xHigh56.8%openai.com for GPT-5.3-Codex (xhigh) on SWE-bench Pro Public (opens in a new tab)Public tasks resolvedReported Launch postOpenAISelf-reportedRecord for GPT-5.3-Codex on SWE-bench Pro Public
Terminal-Bench 2.0Effort: xHigh77.3%openai.com for GPT-5.3-Codex (xhigh) on Terminal-Bench 2.0 (opens in a new tab)Terminal-Bench AccuracyReported Launch postOpenAISelf-reportedRecord for GPT-5.3-Codex on Terminal-Bench 2.0
Terminal-Bench 2.0 — Vals AI Terminus 2Effort: HighConfiguration: Terminus-2 harness64.0%vals.ai for GPT-5.3-Codex (high) on Terminal-Bench 2.0 — Vals AI Terminus 2 (opens in a new tab)Archived copy (opens in a new tab)Terminal-Bench AccuracyReported Benchmark leaderboardVals AIIndependentRecord for GPT-5.3-Codex on Terminal-Bench 2.0 — Vals AI Terminus 2
Vibe Code Bench 1.1 — OpenHands, October 2026 evaluationEffort: xHighConfiguration: OpenHands harness61.8%vals.ai for GPT-5.3-Codex (xhigh) on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Vibe Code Bench ScoreReported Benchmark leaderboardVals AIIndependentRecord for GPT-5.3-Codex on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation