MODEL PROFILE · 模型档案

MiMo-V2.5

xiaomi · mimo-v2.5

Community WeChat Group  |  Discord  |  Telegram  |  Reddit ⚠️ Important: Config Update Notice The config.json and tokenizer_config.json files in this repository have been updated since the initial release. If you downloaded MiMo-V2.5 before this commit (4da2748) , please re-pull or manually update these two files to ensure correct model behavior. Using the outdated config may lead to degraded model performance. We apologize for any inconvenience. Quick fix: hf download XiaomiMiMo/MiMo-V2.5 config.json tokenizer_config.json --local-dir ./MiMo-V2.5 MiMo-V2.5 is a native omnimodal model with strong agentic capabilities, supporting text, image, video, and audio understanding within a unified architecture. Built upon the MiMo-V2-Flash backbone and extended with dedicated vision and audio encoders, it delivers robust performance across multimodal perception, long-context reasoning, and agentic workflows. Key features include:

译文排队中,当前展示来源原文。

文本生成reasoningstructured_outputs个工具
上下文窗口1,050,000 Token
输入 → 输出文本、音频、图像、视频 → 文本
模型权重开放权重
档案更新时间2026/10/29 18:29:04
OVERVIEW

模型介绍

Community WeChat Group  |  Discord  |  Telegram  |  Reddit ⚠️ Important: Config Update Notice The config.json and tokenizer_config.json files in this repository have been updated since the initial release. If you downloaded MiMo-V2.5 before this commit (4da2748) , please re-pull or manually update these two files to ensure correct model behavior. Using the outdated config may lead to degraded model performance. We apologize for any inconvenience. Quick fix: hf download XiaomiMiMo/MiMo-V2.5 config.json tokenizer_config.json --local-dir ./MiMo-V2.5 MiMo-V2.5 is a native omnimodal model with strong agentic capabilities, supporting text, image, video, and audio understanding within a unified architecture. Built upon the MiMo-V2-Flash backbone and extended with dedicated vision and audio encoders, it delivers robust performance across multimodal perception, long-context reasoning, and agentic workflows. Key features include:

主要特点

特点摘要待依据来源补齐,不凭模型名称推断能力。

局限与注意事项

暂无足够来源支持具体缺点结论。请重点验证实际任务表现、时延、数据政策、许可与版本生命周期。

PRICING & COST

来源报价与费用分析

同一模型在不同渠道、地区和配置下可能采用不同价格。下表独立保留各渠道报价,不将渠道价格视为厂商官方统一价格。

来源 / 渠道输入输出计费单位采集时间
OpenRouter · xiaomi/mimo-v2.5 ↗USD 0.4USD 2每百万 Token2026/10/29 17:07:22

部分渠道提供缓存优惠。缓存读取、写入及额外费用详见对应报价来源。

按月估算基础 Token 费用

—

预估不包含缓存优惠、阶梯定价、多模态附加费用、网络检索、税费及推理额外输出。

INDEPENDENT BENCHMARKS

公开评测与运行配置

暂无已采集的可核验公开评分。模型未上榜不代表能力较低。

查看完整已收录榜单 →

常见问题

开放权重是否代表可以免费商用?

不一定。应核对具体版本许可证、商用门槛和使用政策。自部署还会产生算力和维护成本。

这些价格是实时结算价格吗?

不是实时结算价格,而是带采集时间的来源报价快照。请到对应渠道核对当前版本、区域、折扣和计费方式。

资料来源与可追溯记录

许可证:mit。模型介绍与第三方评测均以原来源的具体版本和方法为准。