#129
AN
Claude Haiku 3
Anthropic
34.1 LLMBoard
Anthropic model product
Premium model combining maximum intelligence with practical performance. Best model in the world for coding, agents, and computer use. Most robustly aligned model with best prompt injection resistance of any frontier model. Features extended thinking, 200K context window, 64K max output, and a new effort parameter for controlling reasoning depth. Pricing: $5/$25 per million tokens (input/output).
Updated Aug 10, 2026. Default version: Claude Opus 4.5
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.
No published version under this unique model currently has enough benchmark coverage to calculate a score.
Published benchmark records for the scored version currently unavailable.
No published benchmark result is linked to the scored version.
Preference and agent-evaluation signals from published Arena datasets.
| text factuality | coding | 11 | 1532.3 | 12,608 | N/A | |
| text style control | korean | 13 | 1447.5 | 1,096 | N/A | |
| text | coding | 14 | 1498.3 | 17,181 | N/A | |
| text | instruction following | 16 | 1471.6 | 20,511 | N/A | |
| text factuality | korean | 16 | 1406.6 | 535 | N/A | |
| text factuality | longer query | 16 | 1490.0 | 17,980 | N/A | |
| text style control | longer query | 16 | 1492.2 | 22,452 | N/A | |
| text | longer query | 18 | 1478.8 | 22,452 | N/A | |
| text factuality | german | 18 | 1421.4 | 539 | N/A | |
| text factuality | instruction following | 18 | 1476.4 | 15,920 | N/A | |
| document | overall | 19 | 1461.7 | 7,965 | N/A | |
| document style control | overall | 19 | 1464.3 | 7,965 | N/A | |
| text | hard prompts english | 19 | 1480.6 | 20,042 | N/A | |
| text factuality | expert | 20 | 1504.8 | 3,444 | N/A | |
| text factuality | industry mathematical | 20 | 1472.5 | 2,375 | N/A | |
| text style control | creative writing | 20 | 1460.3 | 10,855 | N/A | |
| text style control | instruction following | 20 | 1474.4 | 20,511 | N/A | |
| text style control | coding | 21 | 1522.9 | 17,181 | N/A | |
| text style control | french | 21 | 1497.8 | 1,864 | N/A | |
| text factuality | math | 22 | 1462.6 | 2,743 | N/A | |
| text factuality | polish | 22 | 1432.0 | 1,218 | N/A | |
| text factuality | spanish | 22 | 1452.6 | 1,262 | N/A | |
| text style control | expert | 22 | 1503.0 | 5,052 | N/A | |
| text style control | industry entertainment and sports and media | 22 | 1452.3 | 13,451 | N/A | |
| text factuality | hard prompts english | 23 | 1502.5 | 15,325 | N/A | |
| text style control | hard prompts | 23 | 1497.6 | 41,175 | N/A | |
| text style control | industry writing and literature and language | 23 | 1464.4 | 16,129 | N/A | |
| text factuality | french | 24 | 1484.4 | 1,279 | N/A | |
| text factuality | industry entertainment and sports and media | 24 | 1442.7 | 9,382 | N/A | |
| text style control | german | 24 | 1470.4 | 1,080 | N/A | |
| text style control | hard prompts english | 24 | 1498.5 | 20,042 | N/A | |
| text style control | multi turn | 24 | 1483.6 | 12,583 | N/A | |
| text style control | spanish | 24 | 1467.9 | 1,898 | N/A | |
| text factuality | multi turn | 25 | 1478.6 | 8,708 | N/A | |
| text | french | 26 | 1477.1 | 1,864 | N/A | |
| text style control | industry legal and government | 26 | 1483.0 | 5,164 | N/A | |
| text | expert | 27 | 1487.2 | 5,052 | N/A | |
| text | hard prompts | 27 | 1475.2 | 41,175 | N/A | |
| text factuality | creative writing | 27 | 1446.8 | 7,237 | N/A | |
| text factuality | industry legal and government | 27 | 1481.9 | 3,398 | N/A | |
| text factuality | industry software and it services | 27 | 1503.9 | 22,427 | N/A | |
| text factuality | hard prompts | 28 | 1490.6 | 40,420 | N/A | |
| text factuality | industry writing and literature and language | 28 | 1452.5 | 11,531 | N/A | |
| text style control | industry business and management and financial operations | 28 | 1471.8 | 13,564 | N/A | |
| text | korean | 29 | 1423.7 | 1,096 | N/A | |
| text style control | industry software and it services | 29 | 1506.5 | 26,095 | N/A | |
| text | english | 30 | 1460.2 | 32,960 | N/A | |
| text | industry software and it services | 30 | 1481.1 | 26,095 | N/A | |
| text factuality | industry business and management and financial operations | 30 | 1467.5 | 9,346 | N/A | |
| text style control | russian | 30 | 1472.8 | 7,567 | N/A | |
| text | multi turn | 31 | 1464.8 | 12,583 | N/A | |
| text factuality | english | 31 | 1471.3 | 32,153 | N/A | |
| text style control | industry medicine and healthcare | 31 | 1486.2 | 4,654 | N/A | |
| text | creative writing | 32 | 1440.9 | 10,855 | N/A | |
| text | industry writing and literature and language | 32 | 1448.6 | 16,129 | N/A | |
| text style control | english | 32 | 1478.3 | 32,960 | N/A | |
| text | industry entertainment and sports and media | 33 | 1431.5 | 13,451 | N/A | |
| text style control | industry life and physical and social science | 33 | 1487.0 | 11,005 | N/A | |
| text style control | japanese | 33 | 1441.4 | 592 | N/A | |
| text | industry business and management and financial operations | 34 | 1447.3 | 13,564 | N/A | |
| text style control | overall | 34 | 1468.8 | 70,504 | N/A | |
| text style control | exclude ties | 34 | 1477.4 | 52,367 | N/A | |
| text | industry mathematical | 36 | 1470.5 | 3,386 | N/A | |
| text factuality | overall | 36 | 1458.2 | 69,447 | N/A | |
| text style control | industry mathematical | 36 | 1478.1 | 3,386 | N/A | |
| webdev | webdev-html | 36 | 1481.3 | 6,432 | N/A | |
| text style control | polish | 37 | 1465.7 | 1,800 | N/A | |
| webdev | overall | 37 | 1467.2 | 12,781 | N/A | |
| webdev | webdev | 37 | 1467.2 | 12,781 | N/A | |
| webdev | webdev-react | 37 | 1455.4 | 6,326 | N/A | |
| text factuality | exclude ties | 38 | 1460.1 | 51,536 | N/A | |
| text factuality | non english | 38 | 1442.7 | 36,788 | N/A | |
| text style control | non english | 38 | 1455.6 | 37,535 | N/A | |
| text | spanish | 39 | 1454.2 | 1,898 | N/A | |
| text factuality | russian | 39 | 1460.8 | 4,840 | N/A | |
| text | overall | 40 | 1449.3 | 70,504 | N/A | |
| text | exclude ties | 40 | 1447.2 | 52,367 | N/A | |
| text | math | 40 | 1459.2 | 4,292 | N/A | |
| text factuality | chinese | 40 | 1492.1 | 2,524 | N/A | |
| text | industry legal and government | 43 | 1458.1 | 5,164 | N/A | |
| text | japanese | 43 | 1412.8 | 592 | N/A | |
| text style control | math | 44 | 1464.7 | 4,292 | N/A | |
| text | russian | 45 | 1446.5 | 7,567 | N/A | |
| text style control | chinese | 45 | 1499.6 | 3,621 | N/A | |
| text | non english | 46 | 1435.2 | 37,535 | N/A | |
| text factuality | industry life and physical and social science | 46 | 1474.4 | 7,401 | N/A | |
| text | german | 50 | 1441.3 | 1,080 | N/A | |
| text factuality | industry medicine and healthcare | 50 | 1472.4 | 3,023 | N/A | |
| text | industry life and physical and social science | 59 | 1452.3 | 11,005 | N/A | |
| text | polish | 60 | 1434.3 | 1,800 | N/A | |
| text | chinese | 68 | 1473.8 | 3,621 | N/A | |
| text | industry medicine and healthcare | 81 | 1441.1 | 4,654 | 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.
| QiHang | claude-opus-4-5-20251101 | global | $0.71 | $3.57 | 200K | |
| Jiekou.AI | claude-opus-4-5-20251101 | global | $4.5 | $22.5 | 200K | |
| NanoGPT | claude-opus-4-5-20251101 | global | $5 | $25 | 200K | |
| LLM Gateway | claude-opus-4-5-20251101 | global | $5 | $25 | 200K | |
| 302.AI | claude-opus-4-5-20251101 | global | $5 | $25 | 200K | |
| Abacus | claude-opus-4-5-20251101 | global | $5 | $25 | 200K | |
| Anthropic | claude-opus-4-5-20251101 | global | $5 | $25 | 200K |
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 4.5 | N/A | N/A | 200K | 64K | No | Proprietary |
A concise description based on the published model registry.
Premium model combining maximum intelligence with practical performance. Best model in the world for coding, agents, and computer use. Most robustly aligned model with best prompt injection resistance of any frontier model. Features extended thinking, 200K context window, 64K max output, and a new effort parameter for controlling reasoning depth. 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.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Claude Opus 4.5.
Claude Opus 4.5's default version was released on Nov 24, 2025.
Claude Opus 4.5'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 $0.71 input and $3.57 output via QiHang.
Claude Opus 4.5 is published under Anthropic in the model registry.
The default version has a 200K token context window.
No. The default version is not marked as having publicly available weights.
7 published provider offerings are linked to the default version.
No nearby ranked alternatives are currently available.