DeepSeek model product
DeepSeek-R1-0528 is the May 28, 2025 version of DeepSeek's reasoning model. It features advanced thinking capabilities and serves as a benchmark comparison for newer models like DeepSeek-V3.1. This model excels in complex reasoning tasks, mathematical problem-solving, and code generation through its thinking mode approach.
Updated Aug 10, 2026. Default version: DeepSeek-R1-0528
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 DeepSeek-R1-0528.
| SimpleQA | 0.9 | 4 | 46 | 93.3% | C | |
| Aider-Polyglot | 0.7 | 6 | 22 | 76.2% | C | |
| AIME 2024 | 0.9 | 7 | 53 | 88.5% | C | |
| MMLU-Redux | 0.9 | 10 | 48 | 80.8% | C | |
| BrowseComp-zh | 0.4 | 13 | 13 | 0.0% | C | |
| CodeForces | 0.6 | 14 | 16 | 13.3% | C | |
| LiveCodeBench | 0.7 | 17 | 73 | 77.8% | C | |
| MMLU-Pro | 0.8 | 17 | 129 | 87.5% | C | |
| HMMT 2025 | 0.8 | 25 | 33 | 25.0% | C | |
| Terminal-Bench | 0.1 | 25 | 25 | 0.0% | C | |
| SWE-bench Multilingual | 0.3 | 34 | 34 | 0.0% | C | |
| AIME 2025 | 0.9 | 55 | 114 | 52.2% | C | |
| BrowseComp | 0.1 | 58 | 58 | 0.0% | C | |
| Humanity's Last Exam | 0.2 | 63 | 92 | 31.9% | C | |
| GPQA | 0.8 | 80 | 233 | 66.0% | C | |
| SWE-Bench Verified | 0.4 | 91 | 104 | 12.6% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text | creative writing | 64 | 1409.5 | 2,427 | N/A | |
| text | polish | 64 | 1430.8 | 1,966 | N/A | |
| text | industry medicine and healthcare | 68 | 1446.6 | 1,242 | N/A | |
| text | japanese | 72 | 1383.8 | 548 | N/A | |
| text | industry entertainment and sports and media | 75 | 1401.7 | 3,203 | N/A | |
| text | russian | 75 | 1425.1 | 1,248 | N/A | |
| text | overall | 79 | 1428.0 | 18,422 | N/A | |
| text | exclude ties | 79 | 1420.3 | 13,274 | N/A | |
| text | non english | 79 | 1415.3 | 9,552 | N/A | |
| text style control | japanese | 80 | 1383.3 | 548 | N/A | |
| text | english | 89 | 1434.3 | 8,864 | N/A | |
| text | industry writing and literature and language | 91 | 1402.9 | 4,222 | N/A | |
| text | chinese | 93 | 1460.2 | 1,005 | N/A | |
| text style control | polish | 96 | 1414.2 | 1,966 | N/A | |
| text style control | russian | 98 | 1414.6 | 1,248 | N/A | |
| text style control | creative writing | 101 | 1393.5 | 2,427 | N/A | |
| text style control | industry medicine and healthcare | 101 | 1443.0 | 1,242 | N/A | |
| text | industry life and physical and social science | 103 | 1430.8 | 3,424 | N/A | |
| text style control | exclude ties | 105 | 1412.9 | 13,274 | N/A | |
| text style control | industry software and it services | 105 | 1455.9 | 5,670 | N/A | |
| text | french | 106 | 1420.2 | 206 | N/A | |
| text style control | overall | 106 | 1422.1 | 18,422 | N/A | |
| text style control | industry entertainment and sports and media | 106 | 1389.7 | 3,203 | N/A | |
| text style control | non english | 106 | 1405.8 | 9,552 | N/A | |
| text | german | 107 | 1397.8 | 560 | N/A | |
| text | industry legal and government | 108 | 1419.8 | 1,388 | N/A | |
| text | industry mathematical | 108 | 1418.9 | 943 | N/A | |
| text | industry software and it services | 108 | 1435.4 | 5,670 | N/A | |
| text | spanish | 108 | 1406.1 | 230 | N/A | |
| text style control | coding | 108 | 1464.4 | 2,725 | N/A | |
| text style control | english | 108 | 1431.7 | 8,864 | N/A | |
| text | korean | 113 | 1353.9 | 295 | N/A | |
| text | industry business and management and financial operations | 115 | 1399.6 | 2,884 | N/A | |
| text style control | hard prompts | 115 | 1433.5 | 6,948 | N/A | |
| text | coding | 116 | 1427.0 | 2,725 | N/A | |
| text | hard prompts | 117 | 1415.9 | 6,948 | N/A | |
| text style control | chinese | 118 | 1446.8 | 1,005 | N/A | |
| text style control | korean | 118 | 1347.7 | 295 | N/A | |
| text style control | french | 120 | 1422.4 | 206 | N/A | |
| text style control | industry mathematical | 120 | 1415.2 | 943 | N/A | |
| text style control | spanish | 120 | 1395.9 | 230 | N/A | |
| text | multi turn | 121 | 1405.3 | 2,954 | N/A | |
| text style control | industry writing and literature and language | 121 | 1390.1 | 4,222 | N/A | |
| text style control | hard prompts english | 122 | 1438.5 | 3,490 | N/A | |
| text | hard prompts english | 124 | 1417.6 | 3,490 | N/A | |
| text style control | german | 126 | 1384.4 | 560 | N/A | |
| text style control | industry legal and government | 127 | 1415.2 | 1,388 | N/A | |
| text style control | industry life and physical and social science | 127 | 1425.6 | 3,424 | N/A | |
| text | math | 128 | 1403.2 | 863 | N/A | |
| text | instruction following | 129 | 1382.0 | 4,046 | N/A | |
| text style control | industry business and management and financial operations | 129 | 1404.2 | 2,884 | N/A | |
| text style control | expert | 131 | 1419.9 | 881 | N/A | |
| text style control | multi turn | 132 | 1405.7 | 2,954 | N/A | |
| text | expert | 134 | 1396.7 | 881 | N/A | |
| text style control | instruction following | 134 | 1391.6 | 4,046 | N/A | |
| text style control | longer query | 137 | 1407.0 | 3,130 | N/A | |
| text | longer query | 138 | 1391.2 | 3,130 | N/A | |
| text style control | math | 143 | 1396.1 | 863 | 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.
| NanoGPT | deepseek-ai/DeepSeek-R1-0528 | global | $0.40 | $1.7 | 128K | |
| submodel | deepseek-ai/DeepSeek-R1-0528 | global | $0.50 | $2.15 | 75K | |
| Deep Infra | deepseek-ai/DeepSeek-R1-0528 | global | $0.50 | $2.15 | 163.8K | |
| Meganova | deepseek-ai/DeepSeek-R1-0528 | global | $0.50 | $2.15 | 163.8K | |
| Alibaba (China) | deepseek-r1-0528 | global | $0.574 | $2.29 | 131.1K | |
| Cortecs | deepseek-r1-0528 | global | $0.652 | $2.57 | 164K | |
| TensorX | deepseek/deepseek-r1-0528 | global | $0.66 | $2.6 | 164K | |
| Kilo Gateway | deepseek/deepseek-r1-0528 | global | $0.70 | $2.5 | 163.8K | |
| NovitaAI | deepseek/deepseek-r1-0528 | global | $0.70 | $2.5 | 163.8K | |
| Jiekou.AI | deepseek/deepseek-r1-0528 | global | $0.70 | $2.5 | 163.8K | |
| Hugging Face | deepseek-ai/DeepSeek-R1-0528 | global | $3 | $5 | 163.8K |
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.
| DeepSeek-R1-0528 | 44.1 | 671B | 131.1K | 131.1K | No | MIT | |
| DeepSeek-R1 | N/A | 671B | 128K | 32.8K | Yes | MIT | |
| DeepSeek R1 Zero | N/A | 671B | N/A | N/A | No | MIT |
A concise description based on the published model registry.
DeepSeek-R1-0528 is the May 28, 2025 version of DeepSeek's reasoning model. It features advanced thinking capabilities and serves as a benchmark comparison for newer models like DeepSeek-V3.1. This model excels in complex reasoning tasks, mathematical problem-solving, and code generation through its thinking mode approach.
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 DeepSeek-R1.
DeepSeek-R1's default version was released on May 28, 2025.
No official standard PAYG price is currently available for DeepSeek-R1. The lowest tracked third-party offer starts at $0.40 input and $1.7 output via NanoGPT.
DeepSeek-R1 is published under DeepSeek in the model registry.
The default version has a 131.1K token context window.
No. The default version is not marked as having publicly available weights.
12 published provider offerings are linked to the default version.
Nearby ranked alternatives include Granite 3.3 8B, MiniMax M2, Llama 3.1 Nemotron Nano 8B.