Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

TOFU - Phi 5% Forget — leaderboard

Metric: Forget Quality x Model Utility. Source: huggingface.co. 3 models tracked.

Top models

#ModelScore
1Phi - Retain Model (WD=0.01)0.53
2Phi - KL Min. (WD=0.01)0.05
3Phi - Grad. Ascent (WD=0.01)0.03

Interactive version: theaggregate.ai/benchmark?slug=tofu-phi-5-forget · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.