Anthropic model product
The most intelligent Claude model and the first hybrid reasoning model on the market. Claude 3.7 Sonnet can produce near-instant responses or extended, step-by-step thinking that is made visible to the user. Shows particularly strong improvements in coding and front-end web development.
Updated Aug 10, 2026. Default version: Claude 3.7 Sonnet
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 3.7 Sonnet.
| TAU-bench Retail | 0.8 | 4 | 25 | 87.5% | C | |
| IFEval | 0.9 | 7 | 65 | 90.6% | C | |
| TAU-bench Airline | 0.6 | 9 | 23 | 63.6% | C | |
| MATH-500 | 1.0 | 14 | 32 | 58.1% | C | |
| Terminal-Bench | 0.4 | 14 | 25 | 45.8% | C | |
| MMMU | 0.8 | 22 | 63 | 66.1% | C | |
| MMMLU | 0.9 | 25 | 49 | 50.0% | C | |
| AIME 2024 | 0.8 | 28 | 53 | 48.1% | C | |
| GPQA | 0.8 | 55 | 233 | 76.7% | C | |
| SWE-Bench Verified | 0.7 | 57 | 104 | 45.6% | C | |
| AIME 2025 | 0.5 | 99 | 114 | 13.3% | C |
Preference and agent-evaluation signals from published Arena datasets.
| vision style control | homework | 68 | 1228.2 | 231 | N/A | |
| vision style control | chinese | 72 | 1175.1 | 153 | N/A | |
| vision | chinese | 74 | 1151.2 | 153 | N/A | |
| vision | homework | 74 | 1206.8 | 231 | N/A | |
| vision style control | diagram | 80 | 1192.4 | 299 | N/A | |
| vision style control | ocr | 81 | 1186.0 | 899 | N/A | |
| vision style control | overall | 85 | 1175.7 | 4,539 | N/A | |
| vision | ocr | 90 | 1158.1 | 899 | N/A | |
| vision | diagram | 91 | 1150.8 | 299 | N/A | |
| vision style control | english | 96 | 1161.3 | 2,196 | N/A | |
| vision | overall | 98 | 1151.0 | 4,539 | N/A | |
| vision | english | 105 | 1140.8 | 2,196 | N/A | |
| text style control | japanese | 117 | 1339.5 | 920 | N/A | |
| text style control | creative writing | 126 | 1376.1 | 5,976 | N/A | |
| text style control | longer query | 131 | 1410.8 | 6,663 | N/A | |
| text style control | industry entertainment and sports and media | 133 | 1371.2 | 7,385 | N/A | |
| text style control | industry writing and literature and language | 139 | 1378.1 | 9,787 | N/A | |
| text style control | korean | 140 | 1329.9 | 674 | N/A | |
| text style control | polish | 140 | 1369.8 | 2,175 | N/A | |
| text style control | multi turn | 142 | 1396.9 | 7,043 | N/A | |
| text style control | instruction following | 143 | 1384.9 | 12,312 | N/A | |
| text style control | german | 145 | 1364.4 | 1,038 | N/A | |
| text | japanese | 149 | 1267.7 | 920 | N/A | |
| text style control | french | 154 | 1388.4 | 505 | N/A | |
| text style control | industry legal and government | 157 | 1394.4 | 2,500 | N/A | |
| text style control | russian | 157 | 1373.0 | 3,193 | N/A | |
| text style control | coding | 160 | 1430.4 | 7,132 | N/A | |
| text style control | hard prompts english | 162 | 1406.1 | 8,820 | N/A | |
| text style control | spanish | 162 | 1361.5 | 560 | N/A | |
| text style control | expert | 163 | 1388.9 | 2,125 | N/A | |
| text style control | hard prompts | 163 | 1398.2 | 15,256 | N/A | |
| text style control | exclude ties | 166 | 1346.5 | 31,697 | N/A | |
| text | longer query | 167 | 1354.8 | 6,663 | N/A | |
| text style control | english | 169 | 1384.9 | 23,477 | N/A | |
| text style control | non english | 169 | 1355.5 | 19,652 | N/A | |
| text style control | overall | 170 | 1371.7 | 43,140 | N/A | |
| text style control | industry life and physical and social science | 170 | 1384.5 | 7,484 | N/A | |
| text style control | industry business and management and financial operations | 171 | 1368.1 | 5,801 | N/A | |
| text style control | industry software and it services | 172 | 1406.4 | 13,015 | N/A | |
| text style control | industry medicine and healthcare | 178 | 1379.4 | 2,264 | N/A | |
| text | creative writing | 179 | 1316.7 | 5,976 | N/A | |
| text | instruction following | 179 | 1325.2 | 12,312 | N/A | |
| text | korean | 179 | 1241.9 | 674 | N/A | |
| text style control | industry mathematical | 179 | 1363.7 | 3,184 | N/A | |
| text style control | math | 179 | 1362.9 | 3,351 | N/A | |
| text style control | chinese | 180 | 1381.1 | 2,196 | N/A | |
| text | industry writing and literature and language | 181 | 1324.1 | 9,787 | N/A | |
| text | french | 183 | 1302.6 | 505 | N/A | |
| text | german | 185 | 1280.2 | 1,038 | N/A | |
| text | multi turn | 185 | 1338.4 | 7,043 | N/A | |
| text | polish | 187 | 1277.3 | 2,175 | N/A | |
| text | spanish | 189 | 1289.1 | 560 | N/A | |
| text | industry entertainment and sports and media | 190 | 1300.4 | 7,385 | N/A | |
| text | math | 194 | 1318.8 | 3,351 | N/A | |
| text | russian | 195 | 1309.8 | 3,193 | N/A | |
| text | expert | 199 | 1300.0 | 2,125 | N/A | |
| text | industry mathematical | 199 | 1313.9 | 3,184 | N/A | |
| text | coding | 201 | 1339.0 | 7,132 | N/A | |
| text | hard prompts | 207 | 1315.2 | 15,256 | N/A | |
| text | non english | 211 | 1285.5 | 19,652 | N/A | |
| text | chinese | 212 | 1298.3 | 2,196 | N/A | |
| text | hard prompts english | 212 | 1322.3 | 8,820 | N/A | |
| text | exclude ties | 217 | 1240.9 | 31,697 | N/A | |
| text | industry software and it services | 217 | 1314.3 | 13,015 | N/A | |
| text | overall | 218 | 1299.4 | 43,140 | N/A | |
| text | industry business and management and financial operations | 221 | 1273.2 | 5,801 | N/A | |
| text | industry legal and government | 222 | 1300.0 | 2,500 | N/A | |
| text | english | 224 | 1311.1 | 23,477 | N/A | |
| text | industry medicine and healthcare | 234 | 1258.5 | 2,264 | N/A | |
| text | industry life and physical and social science | 239 | 1284.5 | 7,484 | 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.
| ZenMux | anthropic/claude-3.7-sonnet | global | $3 | $15 | 200K | |
| Abacus | claude-3-7-sonnet-20250219 | global | $3 | $15 | 200K | |
| DigitalOcean | anthropic-claude-3.7-sonnet | global | $3 | $15 | 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 3.7 Sonnet | 63.6 | N/A | 200K | 64K | No | Proprietary |
A concise description based on the published model registry.
The most intelligent Claude model and the first hybrid reasoning model on the market. Claude 3.7 Sonnet can produce near-instant responses or extended, step-by-step thinking that is made visible to the user. Shows particularly strong improvements in coding and front-end web development.
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 Sonnet 3.7.
Claude Sonnet 3.7's default version was released on Feb 24, 2025.
No official standard PAYG price is currently available for Claude Sonnet 3.7. The lowest tracked third-party offer starts at $3 input and $15 output via ZenMux.
Claude Sonnet 3.7 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.
4 published provider offerings are linked to the default version.
Nearby ranked alternatives include GLM 4.7, Qwen3 235B A22B Thinking, GLM 5.1.