Cohere Parse 5 loses the benchmark on points. It wins on cost per page.

Cohere has released Parse 5, a 2.3-billion-parameter vision language model designed to convert PDFs, slides, and images into structured Markdown for enterprise scale. While Cohere's benchmarks show Parse 5 trails GPT-5.5, Opus 4.8, and Gemini 3.5 Flash in raw accuracy, it excels in cost per page. By prioritizing a price-to-performance balance over absolute accuracy, Parse 5 aims to democratize AI-driven document processing and enable broader industrial adoption by reducing the costs of scaling AI pipelines.