GitHub[195K]Try OpenCode

CompareAI models

  1. DeepSeek V4 FlashDeepSeek
  2. Kimi K2.5Moonshot
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
DeepSeek V4 Flash100/10074/10060/10081/1000/100No data
Kimi K2.5100/10020/10045/10063/10050/100No data
Overview
Author
Context length
1M
262K
Reasoning
True
True
Input modalities
Text
Text, Image, Video
Output modalities
Text
Text
Providers
Pricing
Input
$0.44/ 1M
$0.60/ 1M
Output
$1.32/ 1M
$3.00/ 1M
Cached input
$0.01/ 1M
$0.10/ 1M
MomentumLast 2 mo
Unique users
4.1M
57K
Completed sessions
25,134,417
3,979
Token share
55.4%
0.01%
Tokens
284T+107%
37B-85%
JUL 14SEP 7
JUL 14SEP 7
RetentionWeek 1
Returning users
69.1%455K user-weeks
70.7%834 user-weeks

Related comparisons. Other model pairs to check.