DeepSeek: R1 Distill Qwen 32B
DeepSeek R1 Distill Qwen 32B delivers strong math and science benchmark scores at a very low price, making it appealing for quantitative tasks. However, limited Western API availability and a small 32K context window reduce its practicality for most UK business workflows.
Assessment date: March 12, 2026
Our methodology takes into account a range of factors including pricing, functionality, capabilities, benchmark performance, and real-world applicability. Rankings are reviewed and updated regularly as new models are released. Issues with our rankings? Contact us
DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1. It outperforms OpenAI's o1-mini across various benchmarks, achieving new state-of-the-art results for dense models.nnOther benchmark results include:nn- AIME 2024 pass@1: 72.6n- MATH-500 pass@1: 94.3n- CodeForces Rating: 1691nnThe model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models.
Architecture
| Modality | Text → Text |
| Tokenizer | Qwen |
| Instruct Type | deepseek-r1 |
| Parameters | 32B |
Benchmark Scores
Evaluations
Benchmark data from Artificial Analysis and Hugging Face
Model Information
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.29 | $0.000290 |
| Output | $0.29 | $0.000290 |
Live Performance
Live endpoint metrics — refreshed every 30 minutes.
