GitHub[208K]Try OpenCode

CompareAI models

  1. DeepSeek V4 FlashDeepSeek
  2. Sonar Deep ResearchPerplexityHollow points use fallbacks · hover for details
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
DeepSeek V4 Flash26/10075/10060/10080/1000/1000/100
Sonar Deep Research50/100 — Reasoning supported; no comparable benchmark50/100 — No data; neutral placeholder26/10018/1000/1000/100 — Tool calling not supported
Overview
Author
Context length
1M
128K
Reasoning
True
True
Input modalities
Text
Text
Output modalities
Text
Text
Providers
Pricing
Input
$0.30/ 1M
$2.00/ 1M
Output
$1.20/ 1M
$8.00/ 1M
Cached input
$0.006/ 1M
Unknown
MomentumLast 2 mo
Unique users
4.3M
No usage
Completed sessions
37,642,344
No usage
Token share
32.3%
No usage
Tokens
286T+137%
No usage
AUG 6SEP 30
No trend data
RetentionWeek 1
Returning users
65.3%472K user-weeks
Pending

Related comparisons. Other model pairs to check.