DeepSeek
R1 Distill Qwen 32B
DeepSeek R1 Distill Qwen 32B is a distilled large language model based on [Qwen 2.5 32B](https://huggingface.co/Qwen/Qwen2.5-32B), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). It outperforms OpenAI's o1-mini across various benchmarks, achieving new state-of-the-art results for dense models.\n\nOther benchmark results include:\n\n- AIME 2024 pass@1: 72.6\n- MATH-500 pass@1: 94.3\n- CodeForces Rating: 1691\n\nThe model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models.
- Input / 1M tokens
- $0.290
- Output / 1M tokens
- $0.290
- Context window
- 33K tokens
- Provider
- DeepSeek
- Knowledge cutoff
- 2024-07-31
Performance
Median streaming throughput and first-token latency measured by Artificial Analysis.
- Output tokens / sec
- 43 t/s
- Time to first token
- 0.46s
Benchmarks
Intelligence, coding, and math indexes plus the underlying evaluation scores.
- Intelligence Index
- 17
- Coding Index
- —
- Math Index
- 63
- MMLU-Pro
- 73.9%
- GPQA
- 61.5%
- HLE
- 5.5%
- LiveCodeBench
- 27.0%
- SciCode
- 37.6%
- MATH-500
- 94.1%
- AIME
- 68.7%
Benchmarks via Artificial Analysis