GitHub[150K]Try OpenCode

CompareAI models

  1. Gemini 3.1 Flash LiteGoogle
  2. Kimi K2.6Moonshot
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Gemini 3.1 Flash Lite100/100No data64/10093/100100/100No data
Kimi K2.6100/10081/10041/10061/10060/10033/100
Overview
Author
Context length
1M
262K
Reasoning
True
True
Input modalities
Text, Image, Video, Audio, Pdf
Text, Image, Video
Output modalities
Text
Text
Providers
Pricing
Input
$0.25/ 1M
$0.95/ 1M
Output
$1.50/ 1M
$4.00/ 1M
Cached input
$0.03/ 1M
$0.20/ 1M
MomentumLast 2 mo
Unique users
No usage
332K
Completed sessions
No usage
891,196
Token share
No usage
2.23%
Tokens
No usage
2.9T-33%
No trend data
MAY 20JUL 14

Related comparisons. Other model pairs to check.