GitHub[220K]Try OpenCode

Compare AI models: DeepSeek V4 Flash Vision Exp vs Kimi K3

  1. DeepSeek V4 Flash Vision ExpDeepSeek
  2. Kimi K3MoonshotHollow points use fallbacks · hover for details
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
DeepSeek V4 Flash Vision Exp25/10075/10064/10082/10033/1000/100
Kimi K3100/100100/10027/10093/10050/10050/100 — Tool calling supported; no comparable benchmark
Overview
Author
Context length
1M
1M
Reasoning
True
True
Input modalities
Text, Image
Text, Image, Video
Output modalities
Text
Text
Providers
Pricing
Input
$0.22/ 1M
$1.20/ 1M
Output
$0.65/ 1M
$5.50/ 1M
Cached input
$0.00686/ 1M
$0.20/ 1M
MomentumLast 2 mo
Unique users
276K
248K
Completed sessions
1,197,691
764,952
Token share
0.61%
0.05%
Tokens
6.9T+131%
561B-68%
AUG 30OCT 24
AUG 30OCT 24
RetentionWeek 1
Returning users
63%26K user-weeks
42.1%24K user-weeks

Related comparisons. Other model pairs to check.