← Models
GPT-5.5
OpenAI · proprietary · gpt-5.5
GPT-5.5 is OpenAI's smartest model yet, designed for real work across agentic coding, computer use, knowledge work, and early scientific research. It matches GPT-5.4 per-token latency in real-world serving while reaching a much higher level of intelligence and using significantly fewer tokens to complete the same tasks. GPT-5.5 supports a 1M-token context window in the API and a 400K-token context window in Codex, with state-of-the-art results on Terminal-Bench 2.0, OSWorld-Verified, GDPval, FrontierMath, and CyberGym.
Benchmark scores
| Benchmark | Score |
|---|---|
| Tau2 Telecom | 0.98 |
| ARC-AGI | 0.95 |
| GPQA | 0.94 |
| ARC-AGI v2 | 0.85 |
| BrowseComp | 0.84 |
| MMMU-Pro | 0.83 |
| Terminal-Bench 2.0 | 0.83 |
| CyberGym | 0.82 |
| BixBench | 0.81 |
| OSWorld-Verified | 0.79 |
| MCP Atlas | 0.75 |
| FrontierSWE | 0.73 |
| Humanity's Last Exam | 0.52 |
Pricing
- OpenAI$5.00 / $30.00
Input / output per 1M tokens
AA metrics
No Artificial Analysis link yet.
Arena Elo
- vision1287
- text1476
