Introducing GPT-5.3-Codex 5 Feb 2026
- SWE-Bench Pro (Public) 56.8
- Terminal-Bench 2.0 77.3
- OSWorld-Verified 64.7
- GDPval (wins or ties) 70.9
- SWE-Lancer IC Diamond 81.4
GPT-5.3 Codex
GPT-5.3 Codex ranks first on SWE-Lancer.
46
Benchmarks with a score
1
Ranked first, out of five or more models
5
In the top three
OpenAI Feb 2026
GPT-5.3 Codex
Closed weights Flagship
No Epoch Capabilities Index score yet
- SWE-bench Verified 74.8% #15 /32
- Terminal-Bench 2.0 78.4% #4 /43
- Toolathlon 51.9% #6 /50
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BullshitBench V2 | 24% | #79 of 122 | Official leaderboard | |
| BullshitBench | 14.6% | #88 of 107 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CritPt | 16.9% | #40 of 160 | Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-Lancer | 81.4% | #1 of 10 | Model card | |
| WeirdML | 77.9% | #13 of 112 | Official leaderboard | |
| METR Time Horizon | 5.8 h | #5 of 39 | Epoch AI | |
| ALE-Bench | 1,655 | #17 of 121 | Epoch AI | |
| SlopCodeBench | 26% | #2 of 14 | Official leaderboard | |
| METR Time Horizon 1.1 | 5.8 h | #5 of 26 | Official leaderboard | |
| IOI (Vals) | 43.8% | #12 of 55 | Official leaderboard | |
| LiveSQLBench | 30% | #9 of 36 | Official leaderboard | |
| OpenAI-Proof Q&A | 6.00% | #4 of 9 | Model card | |
| SWE-bench Verified | 74.8% | #15 of 32 | Epoch AI | |
| Vibe Code Bench | 61.8% | #54 of 115 | Official leaderboard | |
| SWE-Bench Pro | 56.8% | #38 of 77 | Model card | |
| IOI v2 (Vals) | 53.8% | #23 of 45 | Official leaderboard | |
| PostTrainBench | 17.8% | #12 of 23 | Official leaderboard | |
| SWE Atlas Refactoring | 42.4% | #9 of 17 | Official leaderboard | |
| SWE Atlas Test Writing | 39% | #12 of 22 | Official leaderboard | |
| SWE Atlas | 32.6% | #14 of 22 | Official leaderboard | |
| Code Arena | 1,407 | #79 of 124 | Official leaderboard | |
| WebDev Arena | 1,370 | #76 of 105 | Epoch AI | |
| Image-to-WebDev Arena | 1,443 | #44 of 55 | Official leaderboard | |
| Monorepo-Bench | 56% | #4 of 5 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 33 | #14 of 183 | Official leaderboard | |
| Terminal-Bench 2.0 | 78.4% | #4 of 43 | Official leaderboard | |
| Toolathlon | 51.9% | #6 of 50 | Official leaderboard | |
| LLM Chess | 1,053 | #23 of 124 | Official leaderboard | |
| OSWorld-Verified | 64.7% | #26 of 37 | Model card | |
| Long-Horizon-Terminal-Bench (LHTB) | 21.5% | #24 of 25 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ProfBench | 57.2% | #10 of 72 | Official leaderboard | |
| prinzbench | 52.0 | #8 of 26 | Official leaderboard | |
| Vending-Bench 2 | $5,940 | #25 of 66 | Official leaderboard | |
| GDPval | 70.9% | #8 of 10 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ArtifactArena | 1,173 | #6 of 21 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ProtocolQA Open-Ended | 44% | #3 of 25 | Model card | |
| Capture the Flag (Professional) | 88% | #5 of 21 | Model card | |
| BinaryAudit | 42.1% | #5 of 20 | Official leaderboard | |
| Destructive action avoidance | 88% | #2 of 8 | Model card | |
| Cyber Range | 80% | #3 of 10 | Model card | |
| CVE-Bench | 90% | #5 of 14 | Model card | |
| Tacit Knowledge and Troubleshooting | 72% | #15 of 29 | Model card | |
| TroubleshootingBench | 32% | #13 of 19 | Model card | |
| Multimodal Troubleshooting Virology | 50% | #19 of 26 | Model card | |
| CyScenarioBench | 0% | #13 of 14 | Model card | |
| Apollo Research sabotage evaluations | 0.88 | #3 of 3 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 15 of them aren't in the launch post.
GPT-5.3-Codex System Card 5 Feb 2026