GitHub[220K]Try OpenCode

Compare AI models: Claude Haiku 5.5 vs DeepSeek V4 Pro

  1. Claude Haiku 5.5Anthropic
  2. DeepSeek V4 ProDeepSeek
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Claude Haiku 5.569/10025/10072/10081/10050/10050/100
DeepSeek V4 Pro100/100100/10046/10081/1000/10068/100
Overview
Author
Context length
1M
1M
Reasoning
True
True
Input modalities
Text, Image, Pdf
Text
Output modalities
Text
Text
Providers
Pricing
Input
$0.10/ 1M
$0.66/ 1M
Output
$0.50/ 1M
$1.98/ 1M
Cached input
$0.01/ 1M
$0.02/ 1M
MomentumLast 2 mo
Unique users
116K
414K
Completed sessions
1,244,499
1,420,191
Token share
0.08%
0.44%
Tokens
971B+100%
5.3T-83%
SEP 3OCT 28
SEP 3OCT 28
RetentionWeek 1
Returning users
59.6%3.2K user-weeks
55%36K user-weeks

Related comparisons. Other model pairs to check.