OpenAI
GPT-5.5
Release date: 4/23/2026
General-purpose frontier intelligence with strong tool use and structured output behavior.
Pramaana IndexAccuracy: StrongLatency: StrongCost: Standard
- Accuracy / Pramaana Index
- 66.80%± 1.33
- Latency / Pramaana Index
- 463.59s
- Cost / Test / Pramaana Index
- $2.63
- Context window
- 512k
- Max output tokens
- 128k
- Input modality
- Text · Image · File
Hyperparameter settings
temperature: 0.0
reasoning: extended
fallbacks: disabled
Benchmark performance for GPT-5.5
| Benchmarks | Accuracy | Ranking |
|---|---|---|
| Pramaana Index | 66.80% ± 1.33 | 6 / 35 |
| Pramaana Multimodal Index | 64.79% ± 1.50 | 4 / 24 |
| Legal Research Fidelity | 61.20% ± 1.67 | 5 / 18 |
| Financial Modeling Agent | 56.91% ± 0.48 | 7 / 22 |
| Clinical Documentation Safety | 60.50% ± 0.65 | 5 / 19 |
| MATH-500 | 78.80% ± 0.82 | 3 / 112 |
| GPQA Diamond | 81.91% ± 0.99 | 3 / 121 |
| MMLU-Pro | 79.40% ± 1.16 | 3 / 119 |
| Adaptive Tutoring Dialogue | 62.40% ± 1.33 | 7 / 16 |
| SWE-bench Verified | 53.31% ± 1.50 | 10 / 69 |
| LiveCodeBench | 72.29% ± 1.67 | 9 / 127 |
| Agentic Reliability | 52.99% ± 0.48 | 4 / 12 |
| Strategic Games Arena | 59.01% ± 0.65 | 6 / 20 |
| Benefits Navigator | 62.31% ± 0.82 | 6 / 16 |