DeepSeek OCR 2

Model updates models.dev · DeepSeek Source language: Chinese

Excerpt

收录发布日期:2026-01-27(社区数据库,待与官方公告核对)。能力:开放权重;输入:text/image。上下文:8,192 tokens。定位(来源描述):High-accuracy OCR model for extracting text from documents, screenshots, receipts, and natural scenes。实际效果:暂无本站独立评测;能力参数不代表评测成绩。资料来自 models.dev 的 Novita AI 目录,厂商目录尚未收录该条目

Short excerpt from the source — open the original for the full text.

Sources

← Back to Daily AI Report