shisa-v2-llama3.1-8B: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1501 ± 30, rank #2336 of 10916 rated models, from 56 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open Japanese LLM - ALT E TO J Bleu JA14.02Score (%)91.1
Open Japanese LLM - ALT E TO J Bert Score JA F186.89Score (%)87.9
Open Japanese LLM - Xlsum JA Bleu JA3.13Score (%)85.1
Open Japanese LLM - Jamp Exact Match66.95Score (%)73.7
Open Japanese LLM - Jsem Exact Match78.03Score (%)63.5
Open Japanese LLM - ALT E TO J Comet Wmt2290.62Score (%)61.4
Open Japanese LLM - RC90.69Score (%)61.4
Open Japanese LLM - Jnli Exact Match83.28Score (%)61.3
Open Japanese LLM - ALT J TO E Bleu EN16.87Score (%)61.2
Open Japanese LLM - Wiki NER SET F16.19Score (%)57.2
JP-TL-Bench - Japanese to English - Hard8.56LT Score (0-10; Hard items only; anchored pairwise Bradley-T57.1
Open Japanese LLM - Wikicorpus E TO J Bleu JA10.71Score (%)56.7

Interactive version: theaggregate.ai/model?slug=shisa-v2-llama3-1-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.