modelgrep

Anthropic: Claude Opus 4.6 vs xAI: Grok 4.3

xAI: Grok 4.3 wins on more metrics (6 of 8), but the right pick depends on what you optimize for — see the breakdown below.

MetricAnthropic: Claude Opus 4.6xAI: Grok 4.3
Intelligence Index52.953.2
Coding Index48.141.0
GPQA Diamond90%90%
Design Arena Elo13581255
Speed (tokens/sec)52150
Latency1.0s638ms
Input price /M$5.00$1.25
Output price /M$25.00$2.50
Context window1M1M
CapabilitiesReasoningToolsJSONVisionReasoningToolsJSONVision