Grok 4
Grok 4 from xAI.
Shogo Agent Index
Supported#68 of 9230.9range 22.4–39.4 · 35% of weight measured
Agentic coding
32
23% measured
Long-horizon autonomy
42
50% measured
Knowledge-work agents
25
48% measured
Computer use
22
estimated
Security & context learning
30
50% measured
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 68 | |
| Terminal-Bench 2.0 | 27.2% | 8% | OpenHands | |
| DeepSWE | est. | 6% | predicted 28 | |
| FrontierSWE | est. | 5% | predicted 5 | |
| CursorBench | est. | 4% | predicted 28 | |
| GSO-Bench | est. | 4% | predicted 11 | |
| METR Time Horizons | 66.6% | 10% | source | |
| Vending-Bench 2 | est. | 6% | predicted 17 | |
| PostTrainBench | est. | 4% | predicted 18 | |
| APEX-Agents | est. | 8% | predicted 28 | |
| GDPval | 21.1% | 7% | source | |
| Remote Labor Index | est. | 5% | predicted 1 | |
| DeepResearch Bench | 47.3% | 5% | True | |
| OSWorld 2.0 | est. | 6% | predicted 3 | |
| OSWorld | est. | 4% | predicted 51 | |
| Cybench | 43.0% | 5% | source | |
| CL-bench | est. | 5% | predicted 18 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| Aider Polyglot | Coding | 79.6% | source |
| ARC-AGI | Reasoning | 66.7% | source |
| ARC-AGI-2 | Reasoning | 16.0% | source |
| BALROG | Long-horizon | 43.6% | source |
| Fiction.LiveBench | Long context | 94.4% | source |
| GeoBench | Multimodal | 45.0% | source |
| GPQA Diamond | Reasoning | 87.0% | source |
| OTIS Mock AIME | Math | 84.0% | source |
| SimpleBench | Reasoning | 60.5% | source |