Anthropic model product
Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).
Updated Aug 10, 2026. Default version: Claude Opus 4.6
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version Claude Opus 4.6.
| Graphwalks parents >128k | 1.0 | 1 | 18 | 100.0% | C | |
| OpenRCA | 0.3 | 1 | 1 | 100.0% | C | |
| Tau2 Retail | 0.9 | 1 | 26 | 100.0% | C | |
| Tau2 Telecom | 1.0 | 1 | 35 | 100.0% | C | |
| Vending-Bench 2 | 8017.6 | 1 | 4 | 100.0% | C | |
| FigQA | 0.8 | 2 | 3 | 50.0% | C | |
| DeepSearchQA | 0.9 | 3 | 8 | 71.4% | C | |
| Finance Agent | 0.6 | 3 | 8 | 71.4% | C | |
| MRCR v2 (8-needle) | 0.8 | 3 | 21 | 90.0% | C | |
| OSWorld | 0.7 | 3 | 20 | 89.5% | C | |
| ARC-AGI v2 | 0.7 | 5 | 16 | 73.3% | C | |
| SWE-bench Multilingual | 0.8 | 5 | 34 | 87.9% | C | |
| AIME 2025 | 1.0 | 6 | 114 | 95.6% | C | |
| CyberGym | 0.7 | 6 | 11 | 50.0% | C | |
| Legal Agent Benchmark | 0.0 | 6 | 13 | 58.3% | B | |
| MMMLU | 0.9 | 6 | 49 | 89.6% | C | |
| SWE-Bench Verified | 0.8 | 7 | 104 | 94.2% | C | |
| FrontierSWE | 0.6 | 8 | 15 | 50.0% | B | |
| LiveBench | 0.8 | 11 | 38 | 73.0% | B | |
| Graphwalks BFS >128k | 0.6 | 14 | 22 | 38.1% | C | |
| Humanity's Last Exam | 0.5 | 14 | 92 | 85.7% | C | |
| BrowseComp | 0.8 | 15 | 58 | 75.4% | C | |
| GPQA | 0.9 | 17 | 233 | 93.1% | C | |
| Terminal-Bench 2.0 | 0.7 | 17 | 49 | 66.7% | C | |
| MCP Atlas | 0.6 | 23 | 30 | 24.1% | C | |
| MMMU-Pro | 0.8 | 23 | 65 | 65.6% | C | |
| CharXiv-R | 0.8 | 27 | 47 | 43.5% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text | spanish | 1 | 1512.0 | 2,292 | N/A | |
| text factuality | spanish | 1 | 1498.5 | 1,652 | N/A | |
| document | overall | 2 | 1509.9 | 37,271 | N/A | |
| text | coding | 2 | 1534.6 | 20,444 | N/A | |
| text | english | 2 | 1506.3 | 34,062 | N/A | |
| text | hard prompts english | 2 | 1527.1 | 22,719 | N/A | |
| text factuality | chinese | 2 | 1544.9 | 3,008 | N/A | |
| text factuality | coding | 2 | 1548.0 | 18,187 | N/A | |
| text factuality | english | 2 | 1499.2 | 33,696 | N/A | |
| text factuality | industry software and it services | 2 | 1533.0 | 27,523 | N/A | |
| text factuality | korean | 2 | 1461.1 | 683 | N/A | |
| text style control | russian | 2 | 1509.3 | 7,658 | N/A | |
| text style control | spanish | 2 | 1502.9 | 2,292 | N/A | |
| text | expert | 3 | 1538.0 | 7,107 | N/A | |
| text | industry business and management and financial operations | 3 | 1499.1 | 14,549 | N/A | |
| text | industry software and it services | 3 | 1527.4 | 28,695 | N/A | |
| text | longer query | 3 | 1516.3 | 30,370 | N/A | |
| text | multi turn | 3 | 1506.9 | 12,925 | N/A | |
| text factuality | expert | 3 | 1537.6 | 5,590 | N/A | |
| text factuality | hard prompts | 3 | 1518.8 | 46,806 | N/A | |
| text factuality | hard prompts english | 3 | 1530.2 | 20,552 | N/A | |
| text factuality | industry mathematical | 3 | 1503.5 | 2,966 | N/A | |
| text factuality | math | 3 | 1497.7 | 2,860 | N/A | |
| text factuality | multi turn | 3 | 1504.0 | 10,451 | N/A | |
| text style control | hard prompts | 3 | 1526.9 | 47,107 | N/A | |
| text style control | hard prompts english | 3 | 1530.3 | 22,719 | N/A | |
| vision | diagram | 3 | 1339.2 | 6,670 | N/A | |
| document style control | overall | 4 | 1494.7 | 37,271 | N/A | |
| text | hard prompts | 4 | 1522.7 | 47,107 | N/A | |
| text | industry entertainment and sports and media | 4 | 1480.3 | 15,598 | N/A | |
| text | industry mathematical | 4 | 1518.8 | 3,900 | N/A | |
| text | instruction following | 4 | 1509.6 | 23,799 | N/A | |
| text | russian | 4 | 1499.7 | 7,658 | N/A | |
| text factuality | overall | 4 | 1487.4 | 72,794 | N/A | |
| text factuality | exclude ties | 4 | 1500.3 | 55,246 | N/A | |
| text factuality | industry entertainment and sports and media | 4 | 1470.7 | 13,567 | N/A | |
| text factuality | instruction following | 4 | 1497.2 | 21,621 | N/A | |
| text factuality | longer query | 4 | 1510.2 | 28,329 | N/A | |
| text factuality | russian | 4 | 1500.4 | 5,829 | N/A | |
| text style control | coding | 4 | 1547.0 | 20,444 | N/A | |
| text style control | english | 4 | 1506.9 | 34,062 | N/A | |
| text style control | expert | 4 | 1535.4 | 7,107 | N/A | |
| text style control | industry business and management and financial operations | 4 | 1502.6 | 14,549 | N/A | |
| text style control | industry entertainment and sports and media | 4 | 1478.1 | 15,598 | N/A | |
| text style control | industry mathematical | 4 | 1516.1 | 3,900 | N/A | |
| text style control | industry software and it services | 4 | 1535.5 | 28,695 | N/A | |
| text style control | industry writing and literature and language | 4 | 1490.6 | 17,487 | N/A | |
| text style control | instruction following | 4 | 1499.1 | 23,799 | N/A | |
| text style control | longer query | 4 | 1518.0 | 30,370 | N/A | |
| vision style control | chinese | 4 | 1358.7 | 1,311 | N/A | |
| vision style control | diagram | 4 | 1330.6 | 6,670 | N/A | |
| vision style control | humor | 4 | 1306.2 | 877 | N/A | |
| text | overall | 5 | 1497.1 | 73,158 | N/A | |
| text | chinese | 5 | 1549.4 | 3,973 | N/A | |
| text | exclude ties | 5 | 1512.3 | 55,526 | N/A | |
| text | industry legal and government | 5 | 1507.4 | 5,819 | N/A | |
| text | industry writing and literature and language | 5 | 1492.7 | 17,487 | N/A | |
| text factuality | industry legal and government | 5 | 1513.3 | 4,451 | N/A | |
| text factuality | industry life and physical and social science | 5 | 1511.8 | 9,650 | N/A | |
| text factuality | industry medicine and healthcare | 5 | 1514.8 | 4,018 | N/A | |
| text factuality | industry writing and literature and language | 5 | 1480.9 | 15,158 | N/A | |
| text factuality | polish | 5 | 1470.7 | 867 | N/A | |
| text style control | overall | 5 | 1497.5 | 73,158 | N/A | |
| text style control | exclude ties | 5 | 1516.0 | 55,526 | N/A | |
| text style control | industry medicine and healthcare | 5 | 1514.0 | 5,320 | N/A | |
| vision | chinese | 5 | 1380.1 | 1,311 | N/A | |
| vision | humor | 5 | 1324.2 | 877 | N/A | |
| vision | ocr | 5 | 1324.3 | 17,878 | N/A | |
| text | industry life and physical and social science | 6 | 1510.6 | 12,007 | N/A | |
| text | industry medicine and healthcare | 6 | 1499.8 | 5,320 | N/A | |
| text | non english | 6 | 1484.7 | 39,095 | N/A | |
| text style control | chinese | 6 | 1548.5 | 3,973 | N/A | |
| text style control | industry life and physical and social science | 6 | 1518.5 | 12,007 | N/A | |
| text style control | multi turn | 6 | 1511.7 | 12,925 | N/A | |
| vision style control | ocr | 6 | 1310.0 | 17,878 | N/A | |
| text | korean | 7 | 1459.4 | 1,319 | N/A | |
| text | polish | 7 | 1488.0 | 1,370 | N/A | |
| text factuality | french | 7 | 1508.0 | 1,808 | N/A | |
| text style control | industry legal and government | 7 | 1506.0 | 5,819 | N/A | |
| vision | overall | 7 | 1311.0 | 24,609 | N/A | |
| vision | english | 7 | 1315.5 | 10,665 | N/A | |
| vision style control | english | 7 | 1292.5 | 10,665 | N/A | |
| text | creative writing | 8 | 1482.4 | 12,085 | N/A | |
| text | french | 8 | 1498.9 | 2,376 | N/A | |
| text | math | 8 | 1509.5 | 3,965 | N/A | |
| text factuality | german | 8 | 1450.7 | 533 | N/A | |
| text factuality | industry business and management and financial operations | 8 | 1493.6 | 12,082 | N/A | |
| text factuality | japanese | 8 | 1469.6 | 474 | N/A | |
| text style control | creative writing | 8 | 1477.8 | 12,085 | N/A | |
| text style control | math | 8 | 1504.9 | 3,965 | N/A | |
| text style control | non english | 8 | 1484.5 | 39,095 | N/A | |
| vision | creative writing vision | 8 | 1319.6 | 1,422 | N/A | |
| text factuality | creative writing | 9 | 1468.5 | 9,770 | N/A | |
| text factuality | non english | 9 | 1474.0 | 38,818 | N/A | |
| vision style control | overall | 9 | 1292.9 | 24,609 | N/A | |
| vision style control | creative writing vision | 9 | 1297.3 | 1,422 | N/A | |
| text | german | 10 | 1488.9 | 1,148 | N/A | |
| text style control | german | 10 | 1494.1 | 1,148 | N/A | |
| text style control | korean | 10 | 1458.5 | 1,319 | N/A | |
| text | japanese | 12 | 1477.7 | 726 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| LLM Gateway | claude-opus-4-6 | global | $5 | $25 | 1M | |
| OpenCode Zen | claude-opus-4-6 | global | $5 | $25 | 1M | |
| 302.AI | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Neon | claude-opus-4-6 | global | $5 | $25 | 1M | |
| FrogBot | claude-opus-4-6 | global | $5 | $25 | 200K | |
| Pioneer | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Abacus | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Azure Cognitive Services | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Auriko | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Azure | claude-opus-4-6 | global | $5 | $25 | 1M | |
| AIHubMix | claude-opus-4-6 | global | $5 | $25 | 1M | |
| FreeModel | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Requesty | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Anthropic | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Jiekou.AI | claude-opus-4-6 | global | $5 | $25 | 1M | |
| Venice AI | claude-opus-4-6 | global | $6 | $30 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Claude Opus 4.6 | 69.2 | N/A | 1M | 128K | No | Proprietary |
A concise description based on the published model registry.
Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Claude Opus 4.6.
Claude Opus 4.6's default version was released on Feb 5, 2026.
Claude Opus 4.6's official API price is $5 per million input tokens and $25 per million output tokens via Anthropic. The lowest tracked third-party offer starts at $5 input and $25 output via LLM Gateway.
Claude Opus 4.6 is published under Anthropic in the model registry.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
16 published provider offerings are linked to the default version.
Nearby ranked alternatives include GPT-5, Llama 3.1 405B, DeepSeek-V2.5.