TAU2
Measured May 14, 2026Source
Score
0.93
Models9 1 
MiMo-V2-Flash (Feb 2026)
XiaomiXiaomiChina
MiMo-V2-Flash is a fast and efficient multimodal AI model from Xiaomi, optimized for on-device and edge deployment. It is the second generation in the MiMo series, focusing on rapid inference and low-latency responses for mobile and IoT applications.
FastMultimodalReasoning
Input / 1M tokens
$0.10Output / 1M tokens
$0.30Output tokens/s
148.6First-token seconds
1.6sSupported plans
4Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.31
Lcr
Measured May 14, 2026Source
Score
0.64
Ifbench
Measured May 14, 2026Source
Score
0.72
Scicode
Measured May 14, 2026Source
Score
0.38
Hle
Measured May 14, 2026Source
Score
0.2
Gpqa
Measured May 14, 2026Source
Score
0.84
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
33.5
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
41.5
Plan availability
Products and plans that support this model
Xiaomi MiMo Token Plan
Xiaomi MiMo Token Plan
Xiaomi MiMo Token Plan is a credit-based subscription service for accessing Xiaomi's MiMo AI models via API, including the flagship MiMo-V2.5-Pro, MiMo-V2.5, and TTS models. It offers tiered credit packages with monthly or annual billing, integrates with popular coding tools, and provides off-peak discounts.
Discussion
Thinking... Make sure you are connected to GitHub server