deepseek-ai/DeepSeek-OCR
by SiliconFlow (China) · listed Oct 2025VW
OCR model for extracting structured text from documents and screenshots
Current rates USD per 1M tokens
Inputfree
Outputfree
Cache read—
Cache write—
Against the market
Input price vs. maker flagships ($/1M tokens)Output price vs. maker flagships ($/1M tokens)On the record
- Model id
deepseek-ai/DeepSeek-OCR- Context window
- 8K tokens
- Max output
- 8K tokens
- Input modalities
- text, image
- Output modalities
- text
- Reasoning
- no
- Tool calls
- no
- Open weights
- yes
- Knowledge cutoff
- —
- Released
- 2025-10-20
- Last updated
- 2025-10-20
← All SiliconFlow (China) listings