GitHub[195K]Try OpenCode

CompareAI models

  1. Qwen3.8 27BAlibaba
  2. Aya Expanse 32BCohere
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
Qwen3.8 27B100/10076/10041/10064/10050/100No data
Aya Expanse 32B0/100No dataNo data19/1000/100No data
Overview
Author
Context length
262K
128K
Reasoning
True
False
Input modalities
Text, Image, Video
Text
Output modalities
Text
Text
Providers
Pricing
Input
$0.55/ 1M
Unknown
Output
$3.30/ 1M
Unknown
Cached input
$0.11/ 1M
Unknown
MomentumLast 2 mo
Unique users
50
No usage
Completed sessions
314
No usage
Token share
0%
No usage
Tokens
186M+100%
No usage
JUL 2AUG 26
No trend data
RetentionWeek 1
Returning users
Pending
Pending

Related comparisons. Other model pairs to check.