Grok 4.5
xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk
Shogo Agent Index
Estimated#36 of 9249.4range 40.9–57.9 · 24% of weight measured
Agentic coding
54
17% measured
Long-horizon autonomy
59
50% measured
Knowledge-work agents
42
32% measured
Computer use
36
estimated
Security & context learning
48
estimated
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | est. | 8% | predicted 78 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 72 | |
| DeepSWE | 53.8% | 6% | mini-swe-agent · $2.42/task | |
| FrontierSWE | est. | 5% | predicted 16 | |
| CursorBench | est. | 4% | predicted 38 | |
| GSO-Bench | est. | 4% | predicted 29 | |
| METR Time Horizons | est. | 10% | predicted 75 | |
| Vending-Bench 2 | $3,887 | 6% | source | |
| PostTrainBench | 22.6% | 4% | Cursor CLI | |
| APEX-Agents | 56.2% | 8% | source | |
| GDPval | est. | 7% | predicted 47 | |
| Remote Labor Index | est. | 5% | predicted 4 | |
| DeepResearch Bench | est. | 5% | predicted 48 | |
| OSWorld 2.0 | est. | 6% | predicted 10 | |
| OSWorld | est. | 4% | predicted 76 | |
| Cybench | est. | 5% | predicted 75 | |
| CL-bench | est. | 5% | predicted 22 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| ARC-AGI | Reasoning | 87.2% | source |
| ARC-AGI-2 | Reasoning | 52.6% | source |
| Chess Puzzles | Reasoning | 35.0% | source |
| CritPt | Reasoning | 15.4% | source |
| DTBench | Knowledge | 96.5% | source |
| FrontierCode | Coding | 42.4% | grok-build |
| FrontierMath Tiers 1-3 | Math | 57.2% | source |
| FrontierMath Tier 4 | Math | 24.4% | source |
| GPQA Diamond | Reasoning | 93.4% | source |
| LMCA | Knowledge | 45.2% | source |
| MirrorCode | Coding | 32.2% | source |
| OTIS Mock AIME | Math | 97.8% | source |
| ProofBench | Math | 31.0% | source |
| SciCode | Coding | 55.0% | source |
| SimpleBench | Reasoning | 70.0% | source |
| SimpleQA Verified | Knowledge | 48.3% | source |
| WebDev Arena | Coding | 1555 | source |