Introducing GPT-5.5 (OpenAI) 23 Apr 2026
GPT-5.5 Pro
GPT-5.5 Pro ranks first on 3 of the 32 benchmarks it has scores on.
OpenAI Released 23 Apr 2026
37
Benchmarks with a score
3
Ranked first, out of five or more models
9
In the top three
OpenAI Apr 2026
GPT-5.5 Pro
Closed weights
163.1
#10 of 271
Epoch Capabilities Index
- CritPt 30.6% #9 /160
- GPQA Diamond 93.9% #14 /219
- FrontierMath 52.4% #1 /70
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 | |
|---|---|---|---|---|
| Humanity's Last Exam | 57.2% | #18 of 147 | Model card | |
| BullshitBench V2 | 36% | #62 of 122 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | #2 of 202 | Epoch AI | |
| FrontierMath | 52.4% | #1 of 70 | Epoch AI | |
| FrontierMath Tier 4 | 39.6% | #2 of 55 | Epoch AI | |
| CritPt | 30.6% | #9 of 160 | Epoch AI | |
| GPQA Diamond | 93.9% | #14 of 219 | Epoch AI | |
| DTBench | 93.3% | #15 of 178 | Official leaderboard | |
| IMO-ProofBench | 88.1% | #1 of 10 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 87.7% | #9 of 89 | Epoch AI | |
| SimpleBench | 76.9% | #11 of 100 | Official leaderboard | |
| ARC-AGI | 96.5% | #11 of 92 | Official leaderboard | |
| ARC-AGI-2 | 84.6% | #13 of 95 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 78% | #10 of 71 | Epoch AI | |
| xbench-ScienceQA | 73% | #7 of 29 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Chess Puzzles | 62% | #7 of 83 | Epoch AI | |
| BrowseComp | 90.1% | #9 of 77 | Model card | |
| PinchBench 2.0 | 21.4% | #55 of 57 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| prinzbench | 82.0 | #2 of 26 | Official leaderboard | |
| CADGenBench | 38.7% | #2 of 11 | Official leaderboard | |
| Investment Banking Modeling Tasks (OpenAI) | 88.6% | #1 of 4 | Model card | |
| GDPval | 82.3% | #4 of 10 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VoxelBench | 1,940 | #13 of 56 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TroubleshootingBench | 49.6% | #1 of 19 | Model card | |
| Agentic BAIM-LLM Evaluation (ABLE) | 83.6% | #2 of 16 | Official leaderboard | |
| CVE-Bench | 93.1% | #2 of 14 | Model card | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 65.2% | #13 of 71 | Official leaderboard | |
| Tacit Knowledge and Troubleshooting | 81.7% | #6 of 29 | Model card | |
| DNA sequence design for transcription factor binding | 16.5% | #1 of 4 | Model card | |
| Multimodal Troubleshooting Virology | 55.3% | #7 of 26 | Model card | |
| ProtocolQA Open-Ended | 36.7% | #11 of 25 | Model card | |
| Biochemistry knowledge improvement | 39.3% | #1 of 2 | Model card | |
| Hard-negative protein binding prediction | 0% | #3 of 4 | Model card | |
| GeneBench | 33.2% | #2 of 2 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Conceptual Reasoning Index | 66.2 | #16 of 149 | Official leaderboard | |
| LMCA | 53.9% | #17 of 139 | Epoch AI | |
| ACCoRD | 75.8 | #24 of 150 | Official leaderboard |
OpenAI's models over time
OpenAI's models in color, against every other model on the Epoch Capabilities Index, by release date.
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. 8 of them aren't in the launch post.
GPT-5.5 System Card 23 Apr 2026
Bio and chem risk
Cyber risk
- CVE-Bench 93.1