Cohere Releases Parse 5 (parse-v5.0): A 2.3B Vision Language Model That Turns Enterprise Documents Into Markdown
Cohere's Parse 5 turns enterprise document chaos into Markdown at $1.50 per 1K pages — a purpose-built model for the PDF-to-structured-data pipeline practitioners have been waiting for.
Why it matters
A specialized vision language model hitting production with clear pricing and benchmark comparisons signals the shift from general-purpose models to task-optimized tools for high-volume document processing — practitioners deploying document pipelines need to evaluate this against their current stack.
The key facts
11 to knowParse 5.0: 2.3B-parameter vision language model
Converts PDFs, slides, images to Markdown with HTML tables, bounding boxes, image descriptions
Pricing: $1.50 per 1,000 pages via API; $2,500/month for dedicated Model Vault instances
ParseBench score: 79.2 (reported as ahead of Mistral OCR 4, Azure Document Intelligence, Databricks AI Parse)
Benchmark caveat: score averages 3 of 5 dimensions; excludes charts and visual grounding
Small model size (2.3B) suggests efficient on-prem or edge deployment possible
Parse 5: 2.3B-parameter vision language model
Pricing: $1.50 per 1,000 pages via API; $2,500/month dedicated instances
ParseBench score: 79.2 (claimed lead over Mistral OCR 4, Azure Document Intelligence, Databricks AI Parse)
Benchmark caveat: averages only 3 of 5 dimensions; excludes charts and visual grounding
Cohere product release (August 2026)
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: Cohere has released Parse (parse-v5.0), a 2.3B-parameter vision language model that converts PDFs, slides and images into Markdown with HTML tables, bounding boxes and image descriptions. It runs at $1.50 per 1,000 pages through the API, or on dedicated Model Vault instances from $2,500 a month.…