Imag-Eval

What is Imag-Eval?

Evaluates text-to-image models on 1,140 prompts with 8,842 combined rules, independently varying instance count and rule constraints to measure compositional…

Released

2026-08-29

Evaluates

General AI, Vision & 3D, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.29210

Imag-Eval paper

• Paper https://arxiv.org/abs/2608.29210

Which sources cite Imag-Eval?

• Imag-Eval

Related General AI, Vision & 3D, cs.AI benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench
• ActiveSciBench