deepseek-r1-distill-qwen-14b
Provider: Alibaba (China), Context: 32768, Output Limit: 16384
| Provider | Source | Input Price ($/1M) | Output Price ($/1M) | Description | Free |
|---|---|---|---|---|---|
| alibabacn | models-dev | Input: $0.14 | Output: $0.43 | Provider: Alibaba (China), Context: 32768, Output Limit: 16384 | |
| siliconflowcn | models-dev | Input: $0.10 | Output: $0.10 | Provider: SiliconFlow (China), Context: 131000, Output Limit: 131000 | |
| siliconflow | models-dev | Input: $0.10 | Output: $0.10 | Provider: SiliconFlow, Context: 131000, Output Limit: 131000 | |
| nscale | litellm | Input: $0.07 | Output: $0.07 | Source: nscale, Context: N/A | |
| fireworksai | litellm | Input: $0.20 | Output: $0.20 | Source: fireworks_ai, Context: 131072 | |
| openrouter | openrouter | Input: $0.15 | Output: $0.15 | DeepSeek R1 Distill Qwen 14B is a distilled large language model based on [Qwen 2.5 14B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B), 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. Other benchmark results include: - AIME 2024 pass@1: 69.7 - MATH-500 pass@1: 93.9 - CodeForces Rating: 1481 The model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models. Context: 32768 |