Claude Opus 5
Strongest Claude Opus model for coding, agents, and professional work
Shogo Agent Index
Supported#11 of 9265.4range 59.8–71 · 39% of weight measured
Agentic coding
74
43% measured
Long-horizon autonomy
74
50% measured
Knowledge-work agents
52
32% measured
Computer use
56
60% measured
Security & context learning
59
estimated
Price
- Input
- $5.00/M
- Output
- $25.00/M
- Cache read
- $0.50/M
- 3:1 blend
- $10.00/M
- Agent-effective
- $2.31/M
Cheapest host for agents: Amazon Bedrock at $2.31/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 | est. | 8% | predicted 87 | |
| Terminal-Bench 2.0 | est. | 8% | predicted 93 | |
| DeepSWE | 73.6% | 6% | mini-swe-agent · $11.84/task | |
| FrontierSWE | 52.0% | 5% | proximus · $196.71/task | |
| CursorBench | 46.6% | 4% | source · $11.95/task | |
| GSO-Bench | est. | 4% | predicted 65 | |
| METR Time Horizons | est. | 10% | predicted 84 | |
| Vending-Bench 2 | $11,182 | 6% | source | |
| PostTrainBench | 35.7% | 4% | Claude Code | |
| APEX-Agents | 65.8% | 8% | source | |
| GDPval | est. | 7% | predicted 65 | |
| Remote Labor Index | est. | 5% | predicted 14 | |
| DeepResearch Bench | est. | 5% | predicted 52 | |
| OSWorld 2.0 | 31.4% | 6% | source | |
| OSWorld | est. | 4% | predicted 93 | |
| Cybench | est. | 5% | predicted 91 | |
| CL-bench | est. | 5% | predicted 28 |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| ARC-AGI | Reasoning | 97.5% | source |
| ARC-AGI-2 | Reasoning | 90.4% | source |
| BALROG | Long-horizon | 63.4% | source |
| Chess Puzzles | Reasoning | 43.0% | source |
| CritPt | Reasoning | 29.1% | source |
| DTBench | Knowledge | 97.9% | source |
| FrontierCode | Coding | 53.4% | claude-code |
| FrontierMath Tiers 1-3 | Math | 85.6% | source |
| FrontierMath Tier 4 | Math | 73.2% | source |
| GPQA Diamond | Reasoning | 93.9% | source |
| LMCA | Knowledge | 64.5% | source |
| MirrorCode | Coding | 87.8% | source |
| OTIS Mock AIME | Math | 98.9% | source |
| ProofBench | Math | 99.0% | source |
| SciCode | Coding | 56.4% | source |
| SimpleBench | Reasoning | 80.6% | source |
| SimpleQA Verified | Knowledge | 59.9% | source |
| WebDev Arena | Coding | 1687 | source |
| WeirdML v3 | Coding | 19.0% | claude_code |
API hosts
Live prices and uptime for every host serving Claude Opus 5 via OpenRouter.
| Host | Input | Output | Cache read | Agent $/M | Uptime 24h | Quant |
|---|---|---|---|---|---|---|
| Amazon Bedrock | $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |
| Anthropic | $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |
| Azure | $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |
| Claude Platform on AWS | $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |
| $5.00 | $25.00 | $0.50 | $2.31 | 99.6% | — | |
| Anthropic | $10.00 | $50.00 | $1.00 | $4.62 | 100.0% | — |