modelgrep

Anthropic: Claude Sonnet 5 vs OpenAI: GPT-5.2-Codex

anthropic logoAnthropic: Claude Sonnet 5anthropic/claude-sonnet-5VSopenai logoOpenAI: GPT-5.2-Codexopenai/gpt-5.2-codex

Anthropic: Claude Sonnet 5 wins on more metrics (7 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricAnthropic: Claude Sonnet 5OpenAI: GPT-5.2-Codex
Intelligence Index53.4✓40.1
Coding Index71.5✓—
GPQA Diamond91%✓90%
Design Arena Elo1330✓—
Speed (tokens/sec)73✓64
Latency2.5s1.6s✓
Input price /M$2.00$1.75✓
Output price /M$10.00✓$14.00
Context window1M✓400K
CapabilitiesReasoningToolsJSONVisionReasoningToolsJSONVision
Anthropic: Claude Sonnet 5 details →OpenAI: GPT-5.2-Codex details →