Comparing 2 models

Claude Sonnet 4.6 vs Grok 4.5

Facts first, then task-specific rankings. A highlighted cell is better only for that row, not universally.

metric
Model 1Claude Sonnet 4.6
Model 2Grok 4.5
Best for
Chat & Roleplay (#4)
AI Agents (#1)
Provider
anthropic
x-ai
Context
1MBest
500K
Input / 1M
$3
$2Best
Output / 1M
$15
$6Best
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
Released
Feb 5, 2026
Jul 7, 2026