Benchmark · Active

IFBench

Name: IFBench, constraints met per prompt, strict: lab-reported scores
Creator: Frontier Log

Created by Ai2 and University of Washington, 3 Jul 2025 · Higher is better · Website

Results

Model	Score	Source	Source check
Nemotron 3.5 Super VL 120B-A12B15 Oct 2026 · NVIDIA	70.86%	nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 model card (revision of 2026-10-15)Our capture	Corroborated The same number is printed in 4 other lab documents, the first 3 linked: 1 2 3
Nemotron 3 Ultra 550B-A55B4 Jun 2026 · NVIDIA	82.94%	Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AIlater report; reported laterOur capture	Corroborated The same number is printed in 1 other lab document: 1
Nemotron 3 Ultra 550B-A55B4 Jun 2026 · NVIDIA	82.9%	nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 model card (revision of 2026-10-15)later report; reported later; another report, not chartedOur capture	Corroborated The same number is printed in 2 other lab documents: 1 2