AnthropicClaude Opus 4.6
Best configuration ranks #48 of 610 on the BenchLeader Index at 63.1 ±3.6. Last measured 8 Sept 2026. Released 5 Feb 2026.
- Blended price
- $10.00/M
- $5.00 in · $25.00 out
- Output speed
- 38 tok/s
- First answer
- 2.05 s
- first token 2.13 s
- Context
- 1M
- Overall index63
- Reasoning62
- Coding62
- Agents & tools69
- Maths66
- Knowledge71
- Instruction following54
- Human preference65
- Multimodal64
- Long context66
- Composite69
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 55.3 | #179 | 38 tok/s | 2.05 s | $0.0095 | Instruction following 52 · Knowledge 66 · Multimodal 56 |
| low | – | – | 38 tok/s | 2.05 s | $0.0095 | Reasoning 63 |
| medium | – | – | 38 tok/s | 2.05 s | $0.0095 | Reasoning 64 |
| high | 61.3 | #69 | 38 tok/s | 2.05 s | $0.0095 | Coding 67 · Composite 50 · Human preference 71 · Maths 61 · Multimodal 68 · Reasoning 67 |
| max | 54.1 | #206 | 38 tok/s | 2.05 s | $0.0095 | Agents & tools 57 · Instruction following 33 · Knowledge 60 · Maths 59 · Multimodal 57 · Reasoning 63 |
| thinking | 62.3 | #59 | 38 tok/s | 2.05 s | $0.0095 | Coding 63 · Knowledge 62 · Maths 65 · Reasoning 75 |
| defaultbest | 63.1 | #48 | 38 tok/s | 2.05 s | $0.0095 | Agents & tools 69 · Coding 62 · Composite 69 · Human preference 65 · Instruction following 54 · Knowledge 71 · Long context 66 · Maths 66 · Multimodal 64 · Reasoning 62 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| GPQA Diamond | – | – | – | – | 88.4%#49 | – | 90.5%#35 | Epoch AI Benchmarking Hub | |
| Humanity's Last Exam | – | – | – | – | 34.4%#8 | – | 19.0%#20 | Scale AI / CAIS | |
| SimpleBench | – | – | – | – | – | – | 67.6%#16 | SimpleBench | |
| LMArena Hard Prompts | – | – | – | 1533#1 | – | – | 1527#3 | LMArena | |
| LiveBench Reasoningnot in index | – | – | – | 88.7%#15 | – | – | – | LiveBench | |
| GPQA Diamond (AA)not in index | – | – | – | – | – | – | 89.6%#71 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | – | – | – | – | – | – | 39.9%#55 | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | – | – | – | – | – | 89.7%#28 | – | Vals AI | |
| Kagi LLM Benchmark | – | – | – | – | – | 83.6%#5 | 72.4%#23 | Kagi LLM Benchmark | |
| ARC-AGI-1 | – | 86.0%#70 | 92.0%#40 | 94.0%#29 | 93.0%#33 | – | – | ARC Prize | |
| ARC-AGI-2 | – | 64.6%#49 | 66.3%#46 | 69.2%#38 | 68.8%#39 | – | – | ARC Prize | |
| ARC-AGI-3 | – | – | – | – | 0.5%#22 | – | – | ARC Prize |
Coding
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| SWE-bench Verified (Epoch) | – | – | – | – | – | – | 78.7%#4 | Epoch AI Benchmarking Hub | |
| WeirdML | – | – | – | 78.0%#18 | – | – | 77.9%#19 | WeirdML | |
| FrontierCode | – | – | – | – | – | – | 26.6%#20 | Cognition | |
| GSO-Bench | – | – | – | 41.2%#4 | – | – | 33.3%#7 | GSO-Bench | |
| LMArena Coding | – | – | – | 1552#2 | – | – | 1546#5 | LMArena | |
| LMArena WebDev | – | – | – | 1546#25 | – | – | 1537#29 | LMArena | |
| LiveBench Codingnot in index | – | – | – | 78.2%#22 | – | – | – | LiveBench | |
| LiveCodeBench | – | – | – | – | – | 84.7%#36 | – | Vals AI | |
| SWE-bench (Vals)not in index | – | – | – | – | – | 78.2%#30 | – | Vals AI | |
| SWE-Bench Pro | – | – | – | – | – | 51.9%#4 | – | Scale AI SEAL | |
| SWE-bench Verified (bash only) | – | – | – | – | – | – | 75.6%#4 | SWE-bench | |
| SWE-bench Verified (any scaffold)not in index | – | – | – | – | – | – | 75.6%#7 | SWE-bench |
Agents & tools
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| Terminal-Bench | – | – | – | – | – | – | 79.8%#5 | Terminal-Bench | |
| Cybench | – | – | – | – | – | – | 93.0%#1 | Cybench | |
| Remote Labor Index | – | – | – | – | – | – | 4.2%#5 | Scale AI / CAIS | |
| APEX-Agents | – | – | – | – | 32.1%#25 | – | 32.4%#24 | Mercor | |
| LiveBench Agentic Codingnot in index | – | – | – | 49.0%#33 | – | – | – | LiveBench | |
| Terminal-Bench Hard | – | – | – | – | – | – | 48.5%#26 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | – | – | – | – | – | – | 92.1%#53 | Artificial Analysis | |
| MCP Atlas | – | – | – | – | 76.8%#14 | – | – | Scale AI SEAL | |
| HiL-Bench | – | – | – | – | – | – | 38.3%#8 | Scale AI SEAL |
Maths
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| FrontierMath Tiers 1–3 | – | – | – | – | 66.0%#21 | – | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | – | – | – | – | 26.8%#31 | – | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | – | – | – | – | 91.1%#55 | – | 94.4%#39 | Epoch AI Benchmarking Hub | |
| ProofBench | – | – | – | – | 50.0%#17 | – | – | Vals AI | |
| LiveBench Mathematicsnot in index | – | – | – | 89.3%#27 | – | – | – | LiveBench | |
| AIME (Vals) | – | – | – | – | – | 95.6%#8 | – | Vals AI | |
| AIME 2026 | – | – | – | 96.7%#6 | – | – | – | MathArena | |
| HMMT February 2026 | – | – | – | 96.2%#5 | – | – | – | MathArena | |
| MathArena Apex | – | – | – | 34.5%#8 | – | – | – | MathArena |
Knowledge
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| SimpleQA Verified | – | – | – | – | 47.0%#29 | – | – | Epoch AI Benchmarking Hub | |
| LiveBench Data Analysisnot in index | – | – | – | 69.9%#41 | – | – | – | LiveBench | |
| AA-Omniscience | – | – | – | – | – | – | 13.7#58 | Artificial Analysis | |
| MMLU-Pro | – | – | – | – | – | 89.1%#15 | – | Vals AI | |
| LegalBench | – | – | – | – | – | 85.3%#19 | – | Vals AI | |
| CorpFin | – | – | – | – | – | 67.0%#13 | – | Vals AI | |
| TaxEval | – | – | – | – | – | 76.0%#8 | – | Vals AI | |
| MedQA | – | – | – | – | – | 95.4%#12 | – | Vals AI | |
| PRBench Finance | 53.3%#3 | – | – | – | – | – | – | Scale AI SEAL | |
| PRBench Legal | 52.3%#4 | – | – | – | – | – | – | Scale AI SEAL | |
| MultiNRC | 48.3%#13 | – | – | – | 57.1%#8 | – | – | Scale AI SEAL |
Instruction following
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LiveBench Languagenot in index | – | – | – | 83.3%#14 | – | – | – | LiveBench | |
| IFBench | – | – | – | – | – | – | 53.1%#155 | Artificial Analysis | |
| MultiChallenge | 56.0%#17 | – | – | – | 37.1%#28 | – | – | Scale AI SEAL | |
| TutorBench | 53.5%#10 | – | – | – | 53.7%#8 | – | – | Scale AI SEAL |
Human preference
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LMArena Text | – | – | – | 1505#2 | – | – | 1498#6 | LMArena | |
| EQ-Bench 4 | – | – | – | – | – | – | 1223#12 | EQ-Bench |
Multimodal
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| LMArena Vision | – | – | – | 1315#5 | – | – | 1311#7 | LMArena | |
| MMMU-Pro | – | – | – | – | – | – | 75.4%#71 | Artificial Analysis | |
| VISTA | 45.5%#25 | – | – | – | 46.1%#22 | – | – | Scale AI SEAL |
Long context
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| AA-LCR | – | – | – | – | – | – | 78.0%#82 | Artificial Analysis |
Composite
| Benchmark | no reasoning | low | medium | high | max | thinking | default | Source | Trend |
|---|---|---|---|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in index | – | – | – | – | – | – | 155.3#22 | Epoch AI Benchmarking Hub | |
| LiveBench | – | – | – | 74.5%#31 | – | – | – | LiveBench | |
| AA Intelligence Index | – | – | – | – | – | – | 31.9#71 | Artificial Analysis |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Amazon Bedrock | 58 tok/s | 1.92 s | $5.00 | $25.00 | 1M | – |
| Google Vertex | 42 tok/s | 2.15 s | $5.00 | $25.00 | 1M | – |
| Anthropic | 34 tok/s | 2.39 s | $5.00 | $25.00 | 1M | – |
| Claude Platform on AWS | 32 tok/s | 1.78 s | $5.00 | $25.00 | 1M | – |
| Azure | 29 tok/s | 2.13 s | $5.00 | $25.00 | 1M | – |
Price history
Listed price per 1M tokens over time, as recorded by OpenRouter for the provider with the longest history.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.500 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0095 | $0.0081 | 10.0 s |
| Summarise a 30-page report | 12,000 / 600 | $0.075 | $0.035 | 17.9 s |
| Code edit | 6,000 / 1,500 | $0.068 | $0.047 | 41.6 s |
| Agentic coding session | 60,000 / 4,000 | $0.400 | $0.198 | 1.8 min |
| Structured extraction | 2,000 / 200 | $0.015 | $0.0083 | 7.3 s |
See also
Data as of 9 Sept 2026. Compare these configurations.