GPT-6.1 Terra
Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.
Shogo Agent Index
Estimated#8 of 9269.0range 60.5–77.5 · 14% of weight measured
Agentic coding
79
estimated
Long-horizon autonomy
74
30% measured
Knowledge-work agents
56
32% measured
Computer use
63
estimated
Security & context learning
62
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 89 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 96 | |
| DeepSWE | est. | 6% | predicted 80 | |
| FrontierSWE | est. | 5% | predicted 58 | |
| CursorBench | est. | 4% | predicted 54 | |
| GSO-Bench | est. | 4% | predicted 75 | |
| METR Time Horizons | est. | 10% | predicted 86 | |
| Vending-Bench 2 | $9,107 | 6% | source | |
| PostTrainBench | est. | 4% | predicted 39 | |
| APEX-Agents | 68.3% | 8% | source | |
| GDPval | est. | 7% | predicted 70 | |
| Remote Labor Index | est. | 5% | predicted 19 | |
| DeepResearch Bench | est. | 5% | predicted 53 | |
| OSWorld 2.0 | est. | 6% | predicted 42 | |
| OSWorld | est. | 4% | predicted 95 | |
| Cybench | est. | 5% | predicted 94 | |
| CL-bench | est. | 5% | predicted 30 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| CritPt | Reasoning | 28.0% | source |
| FrontierMath Tiers 1-3 | Math | 90.9% | source |
| FrontierMath Tier 4 | Math | 92.7% | source |
| GPQA Diamond | Reasoning | 94.3% | source |
| MirrorCode | Coding | 76.7% | source |
| OTIS Mock AIME | Math | 100.0% | source |
| SciCode | Coding | 55.1% | source |
| SimpleBench | Reasoning | 72.2% | source |
| SimpleQA Verified | Knowledge | 57.3% | source |
| WeirdML v3 | Coding | 28.1% | codex_cli |