Loading...
Loading...
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.
Quality Index
17.2
227th of 442
Top 52%
Math Index
63.0
110th of 268
Top 41%
Price/1M
$0.27
323rd cheapest
13% below median
Top 48%
Speed
59 tok/s
Top 43%
TTFT
0.50s
Context Window
33K
291st largest
Top 91%
Input
$0.27
per 1M tokens
Output
$0.27
per 1M tokens
Blended
$0.27
per 1M tokens
Cheaper than 52% of models. Median price is $0.31/1M tokens.
Daily
$0.27
Monthly
$8.10
59
tokens/sec
Faster than 57% of models
0.50
seconds
Faster than 44% of models
34.46
seconds
Faster than 6% of models
Market Median
46 tok/s
29% faster
Median TTFT
0.42s
20% slower
Throughput/Dollar
218
tok/s per $/1M
Speed Comparison
Context Window
33K
tokens
Larger than 9% of models
Max Output
33K
tokens
100% of context
991.6K
1.5K
24-48 GB
A6000 / M3 Ultra