---
格式版本: 2
标题: "Import Xiaomi MiMo V2.5 Pro into OCI Generative AI"
原文链接: "https://docs.oracle.com/en-us/iaas/releasenotes/generative-ai/xiaomi-mimo-v2-5-pro.htm"
发布日期: "2026-08-19"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "llm:scrape:strict_html_body"
发布时间证据: "span class=vl-relnotedate: August 19, 2026（标题附近正文明确标注发布日期）"
发布时间校准原因: "页面为Oracle发布说明，vl-relnotedate位于标题附近，明确标注发布日期为2026-08-19；多个候选一致，c3为正文语义标记，优先采用。修改时间等不作为发布日期。"
发布时间校准置信度: "1"
发布时间候选数量: 16
发布时间严格候选数量: 5
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-21T09:58:27+08:00"
发布时间仲裁状态: "confirmed"
发布时间仲裁尝试次数: 1
发布时间仲裁耗时毫秒: 8012
发现时间: "2026-08-21T09:56:49+08:00"
入库时间: "2026-08-21T01:58:35.843Z"
来源平台: "固定入口"
搜索渠道: "fixed_url"
搜索词: "https://docs.oracle.com/en-us/iaas/releasenotes"
匹配关键词:
  - "AI"
  - "deployment"
相关厂家:
  - "Oracle"
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 18
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "内容为Oracle官方文档，仅介绍小米MiMo模型可导入OCI Generative AI，未涉及超节点/AI Rack机柜级系统、硬件架构、供电散热互连或量产落地等主题，仅提及B200/B300集群单元，无技术展开。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-21T09:58:57+08:00"
AI主题相关性: 2
AI来源权威性: 10
AI新颖性: 0
AI技术细节: 2
AI商业部署信号: 0
AI完整性: 4
AI摘要: "OCI Generative AI 现已支持导入小米 MiMo V2.5 Pro 开源模型并创建端点使用。该 MoE 模型总参数 1.02 万亿、激活参数 420 亿，支持 100 万 token 上下文，面向复杂软件工程与智能体任务；"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T02:13:13.055Z"
采集批次: "2026年8月21日9点50分40秒"
采集批次ID: "20260821-095040-035"
去重键: "https://docs.oracle.com/en-us/iaas/releasenotes/generative-ai/xiaomi-mimo-v2-5-pro.htm"
---

You can now import the [XiaomiMiMo/MiMo-V2.5-Pro](https://docs.oracle.com/iaas/Content/generative-ai/imported-xiaomi-models.htm) model into OCI Generative AI, create endpoints for it, and use it in the Generative AI service.

MiMo-V2.5-Pro is an open source Mixture-of-Experts (MoE) language model with 1.02 trillion total parameters and 42 billion active parameters. Designed for complex software engineering, agentic workflows, and long-horizon tasks, the model supports a context length of up to 1 million tokens and uses hybrid attention and multi-token prediction to improve inference efficiency.

In OCI Generative AI, use the following details for this model:

- **Model capability:** `TEXT_TO_TEXT`
- **Minimum dedicated AI cluster unit shapes:** `B200_X8` and `B300_X8`

For the complete list of models that you can import, see [Models for Import](https://docs.oracle.com/iaas/Content/generative-ai/imported-models.htm). For available hardware and deployment steps, see [Managing Imported Models](https://docs.oracle.com/iaas/Content/generative-ai/manage-imported-models.htm#imported-models).