Alex Getmanearly AI signals
Alex Getman @alexgetmancom · · Updated

Tests & comparisons

Sonnet 5.5 benchmarks are out

🚨 Sonnet 5.5 benchmarks are out

It crushes Sonnet 5, goes toe-to-toe with Opus 5.5, and beats GPT-6 Sol across most shared benchmarks.

Benchmark table comparing Sonnet 5.5, Sonnet 5, Opus 5.5, and GPT-6 Sol across tasks: agentic coding, knowledge work, multidisciplinary reasoning, computer use, and visual chart recognition. Sonnet 5.5 leads in most categories, especially agentic coding and chart recognition.
AnthropicClaude
Русская версияMarkdown

Analysis

Related posts