Claude-Opus-4
Flagship Claude model for deep reasoning, coding, and long-horizon agents
Shogo Agent Index
Supported#75 of 9227.9range 20.4–35.4 · 32% of weight measured
Agentic coding
29
34% measured
Long-horizon autonomy
38
50% measured
Knowledge-work agents
24
20% measured
Computer use
16
estimated
Security & context learning
27
50% measured
Price
- Input
- $13.00/M
- Output
- $64.00/M
- Cache read
- $1.30/M
- 3:1 blend
- $25.75/M
- Agent-effective
- $5.96/M
Agent-effective = 20 input tokens per output token, 85% cache hits. Why.
Agentic benchmarks
| Benchmark | Score | Normalised | Index weight | Harness / source |
|---|---|---|---|---|
| SWE-bench Verified | 70.7% | 8% | source | |
| Terminal-Bench 2.0 | est. | 8% | predicted 26 | |
| DeepSWE | est. | 6% | predicted 20 | |
| FrontierSWE | est. | 5% | predicted 3 | |
| CursorBench | est. | 4% | predicted 24 | |
| GSO-Bench | 6.9% | 4% | OpenHands | |
| METR Time Horizons | 63.9% | 10% | source | |
| Vending-Bench 2 | est. | 6% | predicted 9 | |
| PostTrainBench | est. | 4% | predicted 16 | |
| APEX-Agents | est. | 8% | predicted 20 | |
| GDPval | est. | 7% | predicted 27 | |
| Remote Labor Index | est. | 5% | predicted 1 | |
| DeepResearch Bench | 46.8% | 5% | source | |
| OSWorld 2.0 | est. | 6% | predicted 2 | |
| OSWorld | est. | 4% | predicted 36 | |
| Cybench | 38.0% | 5% | source | |
| CL-bench | est. | 5% | predicted 16 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| Aider Polyglot | Coding | 72.0% | source |
| AlgoTune | Coding | 1.33× | source |
| ARC-AGI | Reasoning | 35.7% | source |
| ARC-AGI-2 | Reasoning | 8.6% | source |
| CritPt | Reasoning | 0.3% | source |
| DTBench | Knowledge | 81.6% | source |
| Fiction.LiveBench | Long context | 61.1% | source |
| GeoBench | Multimodal | 49.0% | source |
| GPQA Diamond | Reasoning | 76.3% | source |
| Humanity's Last Exam | Reasoning | 10.7% | source |
| LMCA | Knowledge | 37.4% | source |
| MATH Level 5 | Math | 85.0% | source |
| OTIS Mock AIME | Math | 64.4% | source |
| SimpleBench | Reasoning | 58.8% | source |
| VPCT | Multimodal | 38.0% | source |