GPT-5 Pro
Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning
Shogo Agent Index
UnrankedNot ranked yet: 0 agentic benchmarks measured, and we need at least 2 plus an overall capability score. It will be ranked as results are published.
Agentic coding
44
estimated
Long-horizon autonomy
50
estimated
Knowledge-work agents
33
estimated
Computer use
30
estimated
Security & context learning
43
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 74 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 58 | |
| DeepSWE | est. | 6% | predicted 39 | |
| FrontierSWE | est. | 5% | predicted 10 | |
| CursorBench | est. | 4% | predicted 33 | |
| GSO-Bench | est. | 4% | predicted 19 | |
| METR Time Horizons | est. | 10% | predicted 71 | |
| Vending-Bench 2 | est. | 6% | predicted 32 | |
| PostTrainBench | est. | 4% | predicted 22 | |
| APEX-Agents | est. | 8% | predicted 37 | |
| GDPval | est. | 7% | predicted 40 | |
| Remote Labor Index | est. | 5% | predicted 3 | |
| DeepResearch Bench | est. | 5% | predicted 47 | |
| OSWorld 2.0 | est. | 6% | predicted 6 | |
| OSWorld | est. | 4% | predicted 66 | |
| Cybench | est. | 5% | predicted 65 | |
| CL-bench | est. | 5% | predicted 20 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| AlgoTune | Coding | 1.31× | source |
| ARC-AGI | Reasoning | 70.2% | source |
| ARC-AGI-2 | Reasoning | 18.3% | source |
| Chess Puzzles | Reasoning | 48.0% | source |
| FrontierMath Tiers 1-3 | Math | 55.8% | source |
| FrontierMath Tier 4 | Math | 19.5% | source |
| Humanity's Last Exam | Reasoning | 31.6% | source |
| SimpleBench | Reasoning | 61.6% | source |
API hosts
Live prices and uptime for every host serving GPT-5 Pro via OpenRouter.
| Host | Input | Output | Cache read | Agent $/M | Uptime 24h | Quant |
|---|---|---|---|---|---|---|
| OpenAI | $15.00 | $120.00 | — | $20.00 | 100.0% | — |