Claude Opus 4.6
High-end Claude for difficult coding, planning, and slower expert reasoning
Shogo Agent Index
Supported#31 of 9253.2range 47.9–58.6 · 64% of weight measured
Agentic coding
57
57% measured
Long-horizon autonomy
67
80% measured
Knowledge-work agents
41
72% measured
Computer use
39
estimated
Security & context learning
57
100% measured
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 | 78.7% | 8% | source | |
| Terminal-Bench 2.0 | 79.8% | 8% | ForgeCode | |
| DeepSWE | est. | 6% | predicted 53 | |
| FrontierSWE | est. | 5% | predicted 19 | |
| CursorBench | est. | 4% | predicted 39 | |
| GSO-Bench | 41.2% | 4% | OpenHands | |
| METR Time Horizons | 78.9% | 10% | source | |
| Vending-Bench 2 | $8,018 | 6% | source | |
| PostTrainBench | est. | 4% | predicted 26 | |
| APEX-Agents | 46.3% | 8% | source | |
| GDPval | est. | 7% | predicted 50 | |
| Remote Labor Index | 4.2% | 5% | source | |
| DeepResearch Bench | 55.3% | 5% | source | |
| OSWorld 2.0 | est. | 6% | predicted 12 | |
| OSWorld | est. | 4% | predicted 80 | |
| Cybench | 93.0% | 5% | source | |
| CL-bench | 20.7% | 5% | source |
Other benchmarks
| Benchmark | Category | Score | Source |
|---|---|---|---|
| AlgoTune | Coding | 1.47× | source |
| ARC-AGI | Reasoning | 94.0% | source |
| ARC-AGI-2 | Reasoning | 69.2% | source |
| Chess Puzzles | Reasoning | 21.0% | source |
| DTBench | Knowledge | 91.2% | source |
| FrontierCode | Coding | 26.6% | claude-code |
| FrontierMath Tiers 1-3 | Math | 66.0% | source |
| FrontierMath Tier 4 | Math | 26.8% | source |
| GPQA Diamond | Reasoning | 90.5% | source |
| Humanity's Last Exam | Reasoning | 34.4% | source |
| LMCA | Knowledge | 55.8% | source |
| MirrorCode | Coding | 33.3% | source |
| OTIS Mock AIME | Math | 94.4% | source |
| ProofBench | Math | 50.0% | source |
| SimpleBench | Reasoning | 67.6% | source |
| SimpleQA Verified | Knowledge | 47.0% | source |
| WebDev Arena | Coding | 1556 | source |
API hosts
Live prices and uptime for every host serving Claude Opus 4.6 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 | 99.9% | — |
| Azure | $5.00 | $25.00 | $0.50 | $2.31 | — | — |
| Claude Platform on AWS | $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |
| $5.00 | $25.00 | $0.50 | $2.31 | 100.0% | — |