GitHub[208K]Try OpenCode

CompareAI models

  1. Claude Haiku 3.5AnthropicHollow points use fallbacks · hover for details
  2. DeepSeek V4.1 FlashDeepSeek
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Claude Haiku 3.50/100 — Reasoning not supported30/10050/100 — No data; neutral placeholder42/10050/10050/100 — Tool calling supported; no comparable benchmark
DeepSeek V4.1 Flash100/100100/10063/10080/10033/10092/100
Overview
Author
Context length
200K
1M
Reasoning
False
True
Input modalities
Text, Image, Pdf
Text, Image
Output modalities
Text
Text
Providers
Pricing
Input
Unknown
$0.27/ 1M
Output
Unknown
$1.08/ 1M
Cached input
Unknown
$0.0054/ 1M
MomentumLast 2 mo
Unique users
0
1.1M
Completed sessions
1
19,447,336
Token share
0%
15%
Tokens
00%
131T+100%
No trend data
AUG 6SEP 30
RetentionWeek 1
Returning users
Pending
58.4%102K user-weeks

Related comparisons. Other model pairs to check.