GitHub[195K]Try OpenCode

CompareAI models

  1. Kimi K2.7 CodeMoonshot
  2. Nemotron 3 Ultra 550B A55BNvidia
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Kimi K2.7 Code100/100No data37/10064/10050/100No data
Nemotron 3 Ultra 550B A55B0/10022/10051/10082/1000/1008/100
Overview
Author
Context length
262K
1M
Reasoning
True
True
Input modalities
Text, Image, Video
Text
Output modalities
Text
Text
Providers
Pricing
Input
$0.95/ 1M
$0.50/ 1M
Output
$4.00/ 1M
$2.20/ 1M
Cached input
$0.19/ 1M
$0.10/ 1M
MomentumLast 2 mo
Unique users
335K
No usage
Completed sessions
537,143
No usage
Token share
0.55%
No usage
Tokens
2.4T+44%
No usage
JUL 7AUG 31
No trend data
RetentionWeek 1
Returning users
61.6%25K user-weeks
Pending

Related comparisons. Other model pairs to check.