MiMo-V2.6-Flash
MiMov2.6 FlashCurrentReleased September 22, 2026
Context1M tokens
Price (In / Out)$0.14/M / $0.28/M
CategoryMultimodal LLM
Max Output128K tokens
About MiMo-V2.6-Flash
MiMo-V2.6-Flash is Xiaomi's MIT-licensed, open-weight multimodal reasoning model optimized for high-frequency and large-scale professional workloads. It supports text, image, video, and audio input, a one-million-token context window, 128K-token output, tool use, web search, structured output, and agent workflows.
Capabilities
reasoningmultimodalcodingtool useweb searchstructured outputcomputer usecontext cachingopen weightslong context
Input Modalities
textimagevideoaudio
Output Modalities
text
Technical Details
- API Identifier
- mimo-v2.6-flash
- Category
- Multimodal LLM
- Context Window
- 1,000,000 tokens
- Max Output Tokens
- 128,000 tokens
Tags
open-weightsreasoningmultimodalcodingagentsefficientlong-context
Benchmarks
Performance scores for MiMo-V2.6-Flash across standard benchmarks.
DeepSWE v1.1xiaomi-mimo · Sep 2026
67.9%
AutomationBench v1.0.6xiaomi-mimo · Sep 2026
52.3%
Toolathlon-Verifiedxiaomi-mimo · Sep 2026
73.6%
Terminal-Bench 4.0xiaomi-mimo · Sep 2026
28.8%
Terminal-Bench 2.1xiaomi-mimo · Sep 2026
87.6%
Artificial Analysis Intelligence Index v4.3.2Artificial Analysis
37.9 index
Pricing
Token pricing for MiMo-V2.6-Flash API usage.
Input Tokens
$0.14/M
per million tokens
Output Tokens
$0.28/M
per million tokens
USD per 1M tokens for Xiaomi MiMo pay-as-you-go API access. Cache-hit input is $0.0028 per 1M tokens.
AI Models by Xiaomi
AI models from the same organization.
| Model | Context Window | Price (In / Out per M) |
|---|---|---|
| MiMo-V2-OmniDeprecated | 262K | Deprecated |
| MiMo-V2-ProDeprecated | 1.0M | Deprecated |
| MiMo-V2-FlashDeprecated | 262K | Deprecated |
Related News
Latest coverage and updates.