Best cell per row highlighted. Null means undisclosed — never counted as zero.
| Provider | OpenAI | OpenAI | OpenAI |
| Family / tier | GPT-5 · Pro | GPT-5 · Pro | GPT-5 · mini |
| Status | GA | GA | GA |
| Released | 2026-04-23 | 2026-04-22 | 2026-03-16 |
| Knowledge cutoff | 2025-12 | 2025-12 | 2025-08 |
| Context window | 1.1M | 1.1M | 400K |
| Max output tokens | 128K | 128K | 128K |
| Input modalities | text, image | text, image | text, image |
| Reasoning mode | always | optional | optional |
| Open weights | No | No | No |
| Input | $30 | $5 | $0.75 |
| Output | $180 | $30 | $4.5 |
| Cached input | — | $0.5 | $0.075 |
| Batch input | $15 | $2.5 | $0.375 |
| Free tier | No | No | No |
| Output speed (tok/s) | — | 68.7 | 180 |
| Time to first token (s) | — | 64 | 0.5 |
| Latency tier | slow | slow | fast |
| Trains on inputs | No | No | No |
| License | Proprietary | Proprietary | Proprietary |
| Clouds | Azure OpenAI, Azure AI Foundry | Azure OpenAI, Azure AI Foundry | Azure OpenAI, Azure AI Foundry |
| Compliance | SOC2, GDPR, CCPA, HIPAA | SOC2, GDPR, CCPA, HIPAA | SOC2, GDPR, CCPA, HIPAA |
| Unified score | 7.7 | 8.7 | 9 |
| Decision Maker | 8 | 9 | 9 |
| Domain Strategist | 7.5 | 8.8 | 9 |
| Finance Lead | 6 | 7 | 10 |
| Domain Practitioner | 7 | 9.2 | 9 |
| Power User | 8 | 9 | 8 |
| Skeptic | 7 | 7.8 | 8.5 |
| Value score | 5 | 6.5 | 9.8 |
| MMLU-Pro | — | 92.4 | — |
| GPQA Diamond | 94 | 93.6 | — |
| SWE-bench Verified | 89 | 88.7 | — |
| MMMU | — | 81.2 | — |
| TAU-bench | — | 98 | — |
| Terminal-Bench | — | 82.7 | — |
| MRCR Long Context | — | 74 | — |
| Humanity's Last Exam | — | 41.4 | — |
| LMArena Elo | — | 1476 | — |
| Artificial Analysis Index | — | 60 | — |
Scores link to their sources. Missing cells mean the vendor hasn't disclosed a result — honesty over padding.