Benchmark RadarAI BENCHMARK PROFILE
CPI-Bench
CPI-Bench evaluates image editing models across general, practical, and intelligent tasks with VLM-as-Judge scoring, including multi-image and reasoning-based editing.
- Released
- 2026-08-14
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
It provides a comprehensive benchmark that captures real-world deployment scenarios and reasoning demands, with scoring aligned to human preferences for reliable model comparison.
Motivation
With the rapid advancement of image editing models and their widespread application across various domains, there is an increasingly urgent need to deploy these model capabilities directly into real-world scenarios.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.