可进入对比
档案摘要 / Decision snapshot
先判断能不能用于选型。
有真实任务证据
来自模型卡
边界优先
30 秒结论
可进入选型对比,但仍以案例证据为准。
来自公开模型资料库,并已补厂商/官方/模型家族证据;当前站内暂无可核验 A 类案例。
这页有来源、有风险说明、有公开可核验 A 类案例。
数据可信度 / 证据强度
强
A 类案例
5
AA 评分
49.2
官方来源
2 个补充入口
案例来源
5 条可核验
来自公开资料与模型数据库 + Artificial Analysis + 2 个厂商/官方/模型家族证据入口;AA 不是唯一事实来源。
Evidence Distribution
案例证据分布
MiMo-V2-Pro 当前关联 5 条 A 类案例;这里按任务、来源和复核时间观察证据结构。
Evidence Snapshot
证据快照覆盖
MiMo-V2-Pro 的 5 条 A 类案例中,已有 5 条完成本地证据快照;0 条仍在快照队列。
可用 manifest 复核
等待落盘
需补抓
需人工复核
证据完整度
5 / 5
Feishu Bitable
2 个入口
5 条
已绑定
已标注
能力边界
用标签表达信号,不用标签替代证据。
通用/待核验
暂无数据
待核验
平台/待核验
中
| 厂商 | Xiaomi / MiMo |
|---|---|
| 厂商页 | Xiaomi / MiMo |
| 发布时间 | 2026-03-18 |
| 模型 ID | mimo-v2-pro |
| 上下文 | 暂无数据 |
| 输出 | 暂无数据 |
| 模态 | 文本 / 多模态待核验 |
| 推理 | 是 / 推理模型或 thinking 模式 |
| 价格 | 官方未披露 / 暂无数据 |
| 可用平台 | Xiaomi / MiMo |
| AA 评分 | 49.2 |
| 基础模型发布时间 | Artificial Analysis 未披露 |
| 资料来源 | 公开资料、厂商信息和案例库 |
| A 类案例 | 5 |
谱系位置
它在厂商路线中的位置
此页将 MiMo-V2-Pro 放入 Xiaomi / MiMo 的当前 Atlas 路线中。若同厂前后代资料不足,先保留为路线追踪入口,不脑补谱系关系。
2026-03-18
MiMo-V2-Pro 资料状态
来自公开资料与模型数据库 + Artificial Analysis + 2 个厂商/官方/模型家族证据入口;AA 不是唯一事实来源。
公开档案
模型状态
MiMo-V2-Pro 当前标记为已有真实案例。
案例库
公开案例复核
已有 5 条可核验 A 类案例,达到完整补齐线
案例时间线
最近进入档案的 A 类案例
用采集时间呈现证据进入 Atlas 的顺序,方便复核来源新鲜度。
2026-06-27T01:35:00Z
OpenClaw user JanPlessow builds mechanical dispatcher agent using MiMo-V2-Pro for tool-call parameter filtering
JanPlessow / agent-tool-calling
2026-06-26T13:59:46Z
Factory AI Coding Agent uses MiMo-V2-Pro for autonomous code generation
Factory-AI / code_generation
2026-06-26T13:59:46Z
MiMo-Code: Xiaomi's terminal-native AI coding assistant powered by MiMo-V2-Pro
Xiaomi MiMo Team / code_generation
2026-06-26T13:59:46Z
ecoSeek Kids: Kid-friendly ecology study chatbot powered by MiMo V2 Pro
alrobles / education_chatbot
2026-06-26T13:59:46Z
MiMo-API-400-Fix: Community compatibility tool enabling MiMo-V2-Pro use with Cursor, Claude Code, and 15+ tools
Miku-cy / developer_tooling
适合场景
- - 作为完整模型索引和厂商路线追踪入口
- - 推理/agentic workflow 候选
不适合
- - 缺少真实案例时不要包装成推荐
- - 缺字段时不要脑补价格、性能或上下文
真实案例
可核验 A 类案例精选
当前展示 5 条代表案例;排序优先看模型卡精选、证据可信度、快照状态和公开仓库/技术任务信号。案例总数仍为 5。
Representative Evidence
代表案例排序
当前第一代表案例是“OpenClaw user JanPlessow builds mechanical dispatcher agent using MiMo-V2-Pro for tool-call parameter filtering”。它会优先出现在本页,是因为该案例同时进入模型卡精选、具备 100/100 的证据可信度,并带有 已快照证据。
模型卡精选优先
A+ 完整链路
archive/evidence/mimo-v2-pro-openclaw-dispatcher-agent/manifest.json
Coding / repo / agent
JanPlessow 使用 MiMo-V2-Pro 处理智能体流程编排
JanPlessow · MiMo-V2-Pro
A+ 完整链路 · 代码仓库证据
JanPlessow 公开的智能体工作流案例,来源为 公开代码库,复核于 2026-06-27T01:35:00Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。
任务:这是一个围绕智能体流程编排的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:JanPlessow built a mechanical dispatcher agent in OpenClaw that calls sessions_spawn to delegate subtasks. They tested openrouter/xiaomi/mimo-v2-pro (alongside Grok 4 Fast) and discovered the model fills in ALL optional…
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:Identified that MiMo-V2-Pro populates all optional tool-schema parameters even when instructed not to, leading to wasted tokens and potential harmful parameter combinations. Proposed and received a config-level fix: per…
模型作用:MiMo-V2-Pro 在该案例中承担智能体流程编排相关的生成、分析、编排或实现角色。 原始资料写作:MiMo-V2-Pro was used as the primary coding model for the dispatcher agent. Its specific behavior (filling all optional schema parameters) was the empirical basis for a new OpenClaw feature — per-agent tool schema parame…
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Model accessed via openrouter/xiaomi/mimo-v2-pro (routed to V2.5-Pro since V2-Pro was deprecated). The behavior observed is from the routed model. Issue is a feature request with working reproduction, not a success stor…
原始记录:OpenClaw user JanPlessow builds mechanical dispatcher agent using MiMo-V2-Pro for tool-call parameter filtering
Miku-cy 使用 MiMo-V2-Pro 处理软件工程任务执行
Miku-cy · MiMo-V2-Pro
A+ 完整链路 · 代码仓库证据
Miku-cy 公开的代码代理与软件工程案例,来源为 公开代码库,复核于 2026-06-26T13:59:46Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。
任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Miku-cy built a compatibility fix tool (API proxy + source patch + config fix) to solve MiMo API's 400 error when assistant messages contain tool_calls without reasoning_content field. The tool enables MiMo-V2-Pro to wo…
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:MIT-licensed v1.1.0 tool with API proxy (auto-fills reasoning_content), OpenClaw source patch, and config fixes for 8 OpenAI-protocol and 7 Anthropic-protocol tools. Includes reasoning_content caching for multi-turn con…
模型作用:MiMo-V2-Pro 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:MiMo-V2-Pro's reasoning/thinking capability is the core value proposition driving this compatibility tool. The 1M context and strong coding performance make it worth the effort to fix API compatibility across 15+ coding…
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Tool was created specifically to fix MiMo API limitations. The need for this tool indicates real-world friction but also strong user demand for MiMo-V2-Pro.
原始记录:MiMo-API-400-Fix: Community compatibility tool enabling MiMo-V2-Pro use with Cursor, Claude Code, and 15+ tools
alrobles 使用 MiMo-V2-Pro 处理医疗和生命科学分析
alrobles · MiMo-V2-Pro
A+ 完整链路 · 代码仓库证据
alrobles 公开的医疗与生命科学案例,来源为 公开代码库,复核于 2026-06-26T13:59:46Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。
任务:这是一个围绕医疗和生命科学分析的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:alrobles built ecoSeek Kids as a kid-friendly (ages 6-15) scientific chatbot focused on ecology, animals, plants, earth science, space, and human body topics. The backend routes all chat through a Hermes Gateway which c…
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:Live production chatbot at kids.ecoseek.org served via Cloudflare Tunnel. Features an animated mascot (Emily Astronauta), topic-specific responses for 6 science domains, and GBIF biodiversity data enrichment. Python ser…
模型作用:MiMo-V2-Pro 在该案例中承担医疗和生命科学分析相关的生成、分析、编排或实现角色。 原始资料写作:MiMo V2 Pro (V2.5 Pro variant) provides the reasoning backbone for science Q&A, enabling the chatbot to generate age-appropriate, factually grounded ecology and nature education content for children. Temperature set to …
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Uses MiMo V2.5 Pro (closely related variant in same V2 family). GitHub issue #9 reports API quota exhaustion (429), confirming active production usage.
原始记录:ecoSeek Kids: Kid-friendly ecology study chatbot powered by MiMo V2 Pro
Factory-AI 使用 MiMo-V2-Pro 处理软件工程任务执行
Factory-AI · MiMo-V2-Pro
A+ 完整链路 · 代码仓库证据
Factory-AI 公开的代码代理与软件工程案例,来源为 公开代码库,复核于 2026-06-26T13:59:46Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。
任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Factory-AI integrated MiMo-V2-Pro as a custom provider in their AI coding agent to autonomously generate, review, and refactor code. The agent uses MiMo-V2-Pro's reasoning/thinking capabilities for complex software engi…
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:Factory AI coding agent running on MiMo-V2-Pro as its backend model, generating code with reasoning traces. The team filed a bug (issue #885) about thinking content being exposed in output, confirming active production …
模型作用:MiMo-V2-Pro 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:MiMo-V2-Pro provides the core reasoning engine for the Factory coding agent, enabling chain-of-thought code generation and complex multi-file refactoring tasks. The model's 1M context window supports large codebase unde…
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Bug report indicates thinking content leaks into output; issue filed June 2026 confirms active usage.
原始记录:Factory AI Coding Agent uses MiMo-V2-Pro for autonomous code generation
Xiaomi MiMo Team 使用 MiMo-V2-Pro 处理多模态内容处理
Xiaomi MiMo Team · MiMo-V2-Pro
A+ 完整链路 · 代码仓库证据
Xiaomi MiMo Team 公开的多模态生成与理解案例,来源为 公开代码库,复核于 2026-06-26T13:59:46Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。
任务:这是一个围绕多模态内容处理的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Xiaomi built MiMo-Code as a terminal-native AI coding assistant that reads/writes code, executes commands, manages Git, and maintains persistent memory across sessions. MiMo-V2-Pro is the default reasoning model behind …
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:MiMo-Code product with 605+ GitHub issues from active users, supporting multiple agents (build/plan/compose), persistent memory via SQLite FTS5, task tracking, subagent system, voice input via MiMo ASR, and Max Mode (pa…
模型作用:MiMo-V2-Pro 在该案例中承担多模态内容处理相关的生成、分析、编排或实现角色。 原始资料写作:MiMo-V2-Pro serves as the core intelligence engine for MiMo-Code's build agent, powering code generation, file editing, command execution, and long-horizon reasoning. The model's 1M context window enables deep project u…
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:605+ issues on GitHub indicate broad adoption. Many issues report loop-thinking and token consumption concerns with mimo-v2-pro.
原始记录:MiMo-Code: Xiaomi's terminal-native AI coding assistant powered by MiMo-V2-Pro
数据缺口
只基于现有字段判断缺口;缺失项不会被猜测填充。
- - 补充方向:核验 价格 的官方/API/案例来源。
- - 补充方向:核验 上下文 的官方/API/案例来源。
- - 补充方向:核验 输出 的官方/API/案例来源。
下一步验证建议
- - 继续保留 A 类案例的原始证据、产物页和版本快照。
- - 补官方发布/API/System Card 来源;若官方未披露,继续标注“官方未披露”。
- - 把 benchmark、教程、发布文保留为背景资料,不提升为真实案例。