Phos-7B: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1455 ± 36, rank #3547 of 10916 rated models, from 50 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Japanese LLM - Xlsum JA RougeLsum24.97Score (%)77.2
Open Japanese LLM - Xlsum JA Rouge129.03Score (%)75.2
Open Japanese LLM - Xlsum JA Rouge210.67Score (%)74.5
Open Japanese LLM - SUM10.67Score (%)74.3
Open Japanese LLM - Xlsum JA Bert Score JA F169.77Score (%)70.7
Open Japanese LLM - Xlsum JA Bleu JA2.53Score (%)44.8
Open Japanese LLM - AIO Char F151.21Score (%)44.5
Open Japanese LLM - ALT E TO J Comet Wmt2289.44Score (%)41.6
Open Japanese LLM - Jamp Exact Match55.17Score (%)38.5
Open Japanese LLM - Commonsensemoralja Exact Match84.64Score (%)37
Open Japanese LLM - Mbpp Pylint Check6.63Score (%)36.5
Open Japanese LLM - Wikicorpus E TO J Comet Wmt2279.17Score (%)35.8

Interactive version: theaggregate.ai/model?slug=phos-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.