Navigate
AI BENCHY
Advertise here

DeepSeek: DeepSeek V3.2 vs Xiaomi: MiMo-V2.5

MiMo-V2.5 leads on average score with 5.1 vs 5.0. MiMo-V2.5 has the lower benchmark cost at $0.025 vs $0.054. MiMo-V2.5 is faster at 4.62s vs 18.25s, with pass rates of 37.9% vs 25.8%.

Recommended modelMiMo-V2.5It has the best score here (5.1), while costing about 2.2x less than DeepSeek V3.2.

Last updated at: 2026-07-22

Metric DeepSeek V3.2 DeepSeek V3.2 none Release: 2025-12-01 MiMo-V2.5 MiMo-V2.5 none Release: 2026-04-22
Score 5.0 5.1
Rank #182 #176
Reliability 10.0 10.0
Consistency 7.7 9.6
Tests Correct
Attempt pass rate 37.9% 25.8%
Flaky tests 6 1
Total Runs 66 66
Cost per result 0.870 0.770
Total Cost $0.054 $0.025
Input Price $0.269 / 1M $0.140 / 1M
Output Price $0.400 / 1M $0.280 / 1M
Total Input Tokens 135,780 141,043
Output Tokens 42,097 16,464
Reasoning Tokens 0 0
Response Time (avg) 18.25s 4.62s
Response Time (max) 115.89s 55.36s
Response Time (total) 401.60s 101.57s

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#182 DeepSeek V3.2

none
Cost
$0.002
Time
7.0s
Tokens
1,046 tok

#176 MiMo-V2.5

none
Cost
$0.007
Time
267.4s
Tokens
25,283 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
DeepSeek V3.2 3.1 6.9 11.1% 1 14.54s 7,279 4,528 0
MiMo-V2.5 5.5 10.0 33.3% 0 3.24s 7,440 696 0

Quick Compare

Switch Comparison Pair