Llama-3.3-FakeSwallow-70B-Instruct-v0.1 Benchmark Results

Llama-3.3-FakeSwallow-70B-Instruct-v0.1: benchmark results

Provider: Meta. Access: Open.

Aggregate Rating 1561 ± 37, rank #1533 of 10916 rated models, from 50 benchmark results.

Benchmark	Score	Metric	Percentile
Open Japanese LLM - Commonsensemoralja Exact Match	93.36	Score (%)	97.8
Open Japanese LLM - AIO Char F1	86.04	Score (%)	95.7
Open Japanese LLM - QA	71.23	Score (%)	95.7
Open Japanese LLM - Wikicorpus E TO J Comet Wmt22	85.36	Score (%)	95.3
Open Japanese LLM - ALT E TO J Bleu JA	14.65	Score (%)	95
Open Japanese LLM - MC	90.84	Score (%)	95
Open Japanese LLM - ALT J TO E Bert Score EN F1	95.91	Score (%)	94.7
Open Japanese LLM - Niilc Char F1	65.96	Score (%)	94.7
Open Japanese LLM - Kuci Exact Match	84.43	Score (%)	94.6
Open Japanese LLM - Wikicorpus E TO J Bert Score JA F1	85.25	Score (%)	94.2
Open Japanese LLM - ALT E TO J Bert Score JA F1	87.19	Score (%)	93.8
Open Japanese LLM - Janli Exact Match	88.33	Score (%)	93.6