MiMo-V2.5
MiMo-V2.5 is a large language model made by Xiaomi, added to the catalog on 22 Apr 2026. It has a context window of 1,000,000 tokens and costs $0.40 per 1M input tokens and $2.00 per 1M output tokens. It accepts text, images, audio and video and produces text. Typical uses: reviewing long documents, image and video understanding and summarising multimodal input.
Price checked ยท source: OpenRouter
What MiMo-V2.5 is
Xiaomi MiMo-V2.5 is a native omnimodal model from Xiaomi. It accepts text, audio, images, and video, and produces text. The listed capabilities include image and video understanding. The model has a context length of 1,000,000 tokens and can produce up to 65,536 tokens in a response, making it suitable for work involving long inputs or extended text generation.
MiMo-V2.5 supports reasoning, tool use, structured outputs, and response-format settings. These options can support tasks that require a response in a specified format or the use of tools as part of a request. Its input and output modalities are distinct: audio, images, and video are accepted alongside text, while text is the listed output. The catalog does not specify particular audio tasks or additional output formats.
The long context can suit reviewing large documents, combining information from lengthy text with image or video inputs, and handling extended conversations. Text generation and structured responses can also suit drafting, extracting information, and producing content in a defined format. These are possible applications based on its listed modalities and supported features; the catalog does not provide further task-specific performance details. MiMo-V2.5 may be considered for projects that need multimodal input, reasoning support, or tool integration within a single model, with text as the response format.
Key features
- Accepts text, audio, image, and video inputs
- Produces text output
- Has a context length of 1,000,000 tokens
- Supports reasoning and tool use
- Supports structured outputs and response-format settings
Typical uses
- Reviewing long documents
- Image and video understanding
- Summarising multimodal input
- Generating structured text responses
- Tool-assisted reasoning tasks
MiMo-V2.5 pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.40 |
| Output, per 1M tokens | $2.00 |
| Cached input read, per 1M tokens | $0.08 |
No price change since it was added on 22 Apr 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0008 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.012 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.016 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.063 |
MiMo-V2.5 specifications
| Model id | xiaomi/mimo-v2.5 |
|---|---|
| Version | mimo-v2.5-20260422 |
| Context window | 1,000,000 tokensAbout 750,000 English words |
| Max output | 65,536 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 22 Apr 2026 |
| Input types | text, audio, image, video |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for MiMo-V2.5 yet.
Compare MiMo-V2.5
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.