GEITje-7B-ultra Benchmark Results · The Aggregate

GEITje-7B-ultra: benchmark results

Provider: Other. Access: Open.

Unified ELO 1358 ± 47, rank #2182 of 2656 rated models, from 19 benchmark results.

Strongest benchmark results

Benchmark	Score	Metric	Percentile
Grip on LLMs - Dutch Simplification - Amsterdam	46.91	SARI (0-100)	100
Grip on LLMs - Dutch Simplification - INT Duidelijke Taal	44.63	SARI (0-100)	100
Grip on LLMs - Dutch Simplification Average	45.77	SARI (0-100; mean of two datasets)	100
Grip on LLMs - HonestCityBench - Narrow Expertise	5	Appropriate acknowledgement rate (%; LLM judge)	91.7
Grip on LLMs - HonestCityBench - Outdated Information	32	Appropriate acknowledgement rate (%; LLM judge)	70
Grip on LLMs - HonestCityBench - Insufficient Information	35	Appropriate acknowledgement rate (%; LLM judge)	66.7
Grip on LLMs - HonestCityBench - Average	26	Appropriate acknowledgement rate (%; mean of five categories	50
Grip on LLMs - HonestCityBench - Non-Text Modality	18	Appropriate acknowledgement rate (%; LLM judge)	48.3
Open LLM Leaderboard - IFEval	37.23	Score	36.1
Grip on LLMs - HonestCityBench - Incorrect Premise	41	Appropriate acknowledgement rate (%; LLM judge)	30
Open LLM Leaderboard - BBH	37.76	Score	21.8
Open LLM Leaderboard - MMLU-Pro	20.11	Score	21.7

Interactive version: theaggregate.ai/model?slug=geitje-7b-ultra · How It Works · Data refreshed daily, snapshot 2026-09-19.