GitHub[195K]Try OpenCode

CompareAI models

  1. Qwen3.8 27BAlibaba
  2. MiMo-V2.5Xiaomi
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Qwen3.8 27B100/10076/10042/10064/10050/100No data
MiMo-V2.5100/100No data82/10093/10067/100No data
Overview
Author
Context length
262K
1M
Reasoning
True
True
Input modalities
Text, Image, Video
Text, Image, Audio, Video
Output modalities
Text
Text
Providers
Pricing
Input
$0.55/ 1M
$0.14/ 1M
Output
$3.30/ 1M
$0.28/ 1M
Cached input
$0.11/ 1M
$0.003/ 1M
MomentumLast 2 mo
Unique users
50
911K
Completed sessions
314
8,672,347
Token share
0%
6.44%
Tokens
186M+100%
26T+167%
JUL 2AUG 26
JUL 2AUG 26

Related comparisons. Other model pairs to check.