Editor's Synopsis
Contra shares results from a benchmark test pitting four leading AI models against evaluations by real professional designers. The models tested include Anthropic's Claude Opus 4.6, Google's Gemini 3.1 Pro, OpenAI's ChatGPT 5.3 Codex, and Alibaba's Qwen 3.5. The post teases that the results "might surprise you" and links to an accompanying video or detailed breakdown.
This comparison is notable for its focus on design-oriented tasks evaluated by working professionals rather than standard automated benchmarks, potentially offering more practical insight into how these AI models perform in creative and professional workflows.