OpenAI · model
GPT-5.3-Codex-Spark benchmark scores
As of 2026-10-29, GPT-5.3-Codex-Spark (OpenAI) has measured scores on 1 benchmark and estimated scores on 12 more.
| Benchmark | Score | Source |
|---|---|---|
| BrowseComp | 77.6% | estimated ± 11.3 pp, low confidence |
| HLE w/ tools | 53.7% | estimated ± 7.5 pp, low confidence |
| ApprenticeBench | 12.9% | estimated ± 12.5 pp, low confidence |
| MazeBench | 0.0% | measured |
| CWE-Bench-AA | 44.0% | estimated ± 11.3 pp, low confidence |
| GDP.pdf | 21.0% | estimated ± 5.1 pp, low confidence |
| CWE-bench v1 | 57.3% | estimated ± 5.4 pp, medium confidence |
| OSWorld 2.1 | 64.8% | estimated ± 7.8 pp, medium confidence |
| Toolathlon-Verified | 73.4% | estimated ± 2.6 pp, high confidence |
| Hermes Index | 37.2% | estimated ± 3.1 pp, high confidence |
| ExploitGym | 12.0% | estimated ± 4.0 pp, high confidence |
| Finance Agent v2 | 56.1% | estimated ± 3.8 pp, medium confidence |
| Agents' Last Exam | 35.2% | estimated ± 10.7 pp, low confidence |