ABACUS Benchmark

What is ABACUS?

Evaluates instruction-based image editing models on add, remove, and change tasks, measuring exact object count correctness and scene preservation without VLM…

Released

2026-09-09

Evaluates

General AI, Vision & 3D, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://huggingface.co/datasets/Aymen-Lassoued/ABACUS_Bench

ABACUS dataset

• Dataset https://huggingface.co/datasets/Aymen-Lassoued/ABACUS_Bench

Which sources cite ABACUS?

• ABACUS

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

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 3DCodeBench
• 5-Dialects-BN
• A Real-World CAD Segmentation Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench