GitHub[208K]Try OpenCode

CompareAI models

  1. Kimi K2.7 CodeMoonshot
  2. Qwen3.8 FlashAlibaba
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Kimi K2.7 Code100/100No data37/10062/10050/100No data
Qwen3.8 Flash100/100No data76/10081/10050/100No data
Overview
Author
Context length
262K
1M
Reasoning
True
True
Input modalities
Text, Image, Video
Text, Image, Video
Output modalities
Text
Text
Providers
Pricing
Input
$0.95/ 1M
$0.15/ 1M
Output
$4.00/ 1M
$0.47/ 1M
Cached input
$0.19/ 1M
$0.02/ 1M
MomentumLast 2 mo
Unique users
264K
99K
Completed sessions
411,613
606,460
Token share
0.27%
0.22%
Tokens
1.8T-35%
1.4T+100%
JUL 25SEP 18
JUL 25SEP 18
RetentionWeek 1
Returning users
60.3%21K user-weeks
72.3%2.2K user-weeks

Related comparisons. Other model pairs to check.