Gemini 3 Deep Think
Gemini 3 Deep Think from Google.
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
70
estimated
Long-horizon autonomy
72
estimated
Knowledge-work agents
50
estimated
Computer use
52
estimated
Security & context learning
58
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 85 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 91 | |
| DeepSWE | est. | 6% | predicted 70 | |
| FrontierSWE | est. | 5% | predicted 39 | |
| CursorBench | est. | 4% | predicted 48 | |
| GSO-Bench | est. | 4% | predicted 58 | |
| METR Time Horizons | est. | 10% | predicted 83 | |
| Vending-Bench 2 | est. | 6% | predicted 81 | |
| PostTrainBench | est. | 4% | predicted 33 | |
| APEX-Agents | est. | 8% | predicted 63 | |
| GDPval | est. | 7% | predicted 62 | |
| Remote Labor Index | est. | 5% | predicted 11 | |
| DeepResearch Bench | est. | 5% | predicted 51 | |
| OSWorld 2.0 | est. | 6% | predicted 26 | |
| OSWorld | est. | 4% | predicted 91 | |
| Cybench | est. | 5% | predicted 89 | |
| CL-bench | est. | 5% | predicted 27 |