Loading...
Loading...
MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting hybrid attention architecture. MiMo-V2-Flash supports a hybrid-thinking toggle and a 256K context window, and excels at reasoning, coding, and agent scenarios. On SWE-bench Verified and SWE-bench Multilingual, MiMo-V2-Flash ranks as the top #1 open-source model globally, delivering performance comparable to Claude Sonnet 4.5 while costing only about 3.5% as much. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://openrouter.ai/docs/use-cases/reasoning-tokens#enable-reasoning-with-default-config).
Quality Index
30.4
100th of 444
Top 23%
Coding Index
25.8
102nd of 354
Top 29%
Math Index
67.7
101st of 268
Top 38%
Price/1M
$0.15
261st cheapest
50% below median
Top 40%
Speed
132 tok/s
Top 18%
TTFT
1.33s
Context Window
262K
61st largest
Top 25%
Input
$0.10
per 1M tokens
Output
$0.30
per 1M tokens
Blended
$0.15
per 1M tokens
Cheaper than 60% of models. Median price is $0.30/1M tokens.
Daily
$0.15
Monthly
$4.50
132
tokens/sec
Faster than 82% of models
1.33
seconds
Faster than 21% of models
1.33
seconds
Faster than 33% of models
Market Median
45 tok/s
192% faster
Median TTFT
0.42s
218% slower
Throughput/Dollar
883
tok/s per $/1M
Speed Comparison
Context Window
262K
tokens
Larger than 75% of models
Max Output
66K
tokens
25% of context