shisa-v2.1-llama3.3-70B: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1620 ± 52, rank #838 of 10916 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| JP-TL-Bench - English to Japanese - Hard | 9.79 | LT Score (0-10; Hard items only; anchored pairwise Bradley-T | 87.3 |
| JP-TL-Bench - English to Japanese | 9.63 | LT Score (0-10; anchored pairwise Bradley-Terry vs 20 Base S | 81.7 |
| JP-TL-Bench - Japanese to English - Hard | 9.8 | LT Score (0-10; Hard items only; anchored pairwise Bradley-T | 79.4 |
| JP-TL-Bench - Japanese to English | 9.66 | LT Score (0-10; anchored pairwise Bradley-Terry vs 20 Base S | 77 |
| JP-TL-Bench - English to Japanese - Easy | 9.5 | LT Score (0-10; Easy items only; anchored pairwise Bradley-T | 76.2 |
| JP-TL-Bench - Japanese to English - Easy | 9.43 | LT Score (0-10; Easy items only; anchored pairwise Bradley-T | 69.8 |
| Open Japanese LLM v2 - JamC-QA | 46.38 | Exact Match (%) | 29.2 |
| Open Japanese LLM v2 - M-IFEval-Ja | 35.47 | Prompt-Level Strict Accuracy (%) | 23.1 |
Interactive version: theaggregate.ai/model?slug=shisa-v2-1-llama3-3-70b · How It Works · Data refreshed daily, snapshot 2026-11-08.