Compare/Kimi K2.5 (Non-reasoning) vs MiMo-V2-Pro

Kimi K2.5 (Non-reasoning)vsMiMo-V2-Pro

Side-by-side comparison of pricing, 12 benchmarks, and generation speed.

Kimi

Kimi K2.5 (Non-reasoning)

Input
$0.6/M
Output
$3/M
Speed
33 tok/s
TTFT
1.40s
Xiaomi

MiMo-V2-Pro

Input
$1/M
Output
$3/M
Speed
83 tok/s
TTFT
1.55s

Winner by Category

Cheaper
Kimi K2.5 (Non-reasoning)
Faster (tok/s)
MiMo-V2-Pro
Lower Latency
Kimi K2.5 (Non-reasoning)
Benchmarks (0-7)
MiMo-V2-Pro

Pricing Comparison

MetricKimi K2.5 (Non-reasoning)MiMo-V2-Pro
Input ($/M tokens)$0.6$1
Output ($/M tokens)$3$3
Cost for 1M input + 100K output tokens:
Kimi K2.5 (Non-reasoning)$0.90
MiMo-V2-Pro$1.30

Speed Comparison

Output Speed (tokens/s) — higher is better
Kimi K2.5 (Non-reasoning)
33 tok/s
MiMo-V2-Pro
83 tok/s
Time to First Token (seconds) — lower is better
Kimi K2.5 (Non-reasoning)
1.40s
MiMo-V2-Pro
1.55s

Benchmark Comparison

Data from Artificial Analysis API — 12 benchmarks

Intelligence Index
37.349.2
Coding Index
25.841.4
Math Index
GPQA Diamond
78.9%87.0%
MMLU-Pro
LiveCodeBench
AIME 2025
MATH-500
Humanity's Last Exam
12.3%28.3%
SciCode
39.6%42.5%
IFBench
43.7%68.8%
TerminalBench
18.9%40.9%
Kimi K2.5 (Non-reasoning)0 wins
7 winsMiMo-V2-Pro

Frequently Asked Questions

Which is cheaper, Kimi K2.5 (Non-reasoning) or MiMo-V2-Pro?

Kimi K2.5 (Non-reasoning) is cheaper overall. Its blended price (3:1 input/output ratio) is $1.20/M tokens vs $1.50/M for MiMo-V2-Pro.

Which model performs better on benchmarks?

MiMo-V2-Pro wins 7 out of 12 benchmarks compared to 0 for Kimi K2.5 (Non-reasoning). See the detailed benchmark chart above for per-category results.

Which is faster for real-time applications?

MiMo-V2-Pro generates tokens faster at 83 tok/s vs 33 tok/s. Kimi K2.5 (Non-reasoning) also has lower time-to-first-token (1.40s vs 1.55s).

When should I use Kimi K2.5 (Non-reasoning) vs MiMo-V2-Pro?

Choose based on your priorities: Kimi K2.5 (Non-reasoning) for lower cost, MiMo-V2-Pro for stronger benchmark performance, and MiMo-V2-Pro for faster generation. For latency-sensitive apps, check the TTFT comparison above.