About
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...
Related Models
Pricing
Input
$0.10
per 1M tokens
Output
$0.30
per 1M tokens
Blended
$0.15
per 1M tokens
Cheaper than 75% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.15
Monthly
$4.50
vs. Similar Models
DeepSeek: DeepSeek V3.2Q:0.0
$0.26+72%
Gemma 4 31B (Non-reasoning)Q:+0.1
$0.20+37%
Grok 4.3 (Non-reasoning)Q:+0.1
$1.56+942%
Arcee AI: Trinity Large ThinkingQ:-0.2
$0.39+158%
Performance
92
tokens/sec
Faster than 49% of models
1.88
seconds
Faster than 24% of models
1.88
seconds
Faster than 51% of models
Market Median
94 tok/s
2% slower
Median TTFT
1.10s
70% slower
Throughput/Dollar
614
tok/s per $/1M
Speed Comparison
Hermes 4 - Llama-3.1 70B (Reasoning)
92 tok/s+0%
Qwen3 32B (Non-reasoning)
91 tok/s-1%
Reka Flash 3
93 tok/s+1%
Context Window
262K
tokens
Larger than 62% of models
Max Output
66K
tokens
25% of context
Benchmarks
MMLU-Pro
74.4%
GPQA Diamond
65.6%
HLE
8.0%
LiveCodeBench
40.2%
SciCode
25.9%
TerminalBench Hard
25.8%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
67.7%
IFBench
39.9%
Long Context Recall
31.3%
Tau2
83.9%
Market AverageTop Score