Llama-3.1-Swallow-70B-Instruct-v0.3: benchmark results
Provider: Institute of Science Tokyo. Released 2024-12-30. Access: Open.
Aggregate Rating 1578 ± 27, rank #1270 of 10916 rated models, from 58 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| pfgen-bench - Completion Mode - Truthfulness | 0.94 | Truthfulness Score | 99.2 |
| Open Japanese LLM - ALT E TO J Comet Wmt22 | 91.56 | Score (%) | 97.8 |
| Open Japanese LLM - MT | 85.9 | Score (%) | 97.2 |
| Open Japanese LLM - Wikicorpus J TO E Comet Wmt22 | 77.13 | Score (%) | 96.6 |
| Open Japanese LLM - Wiki Reading Char F1 | 90.34 | Score (%) | 96.2 |
| Open Japanese LLM - Wikicorpus E TO J Comet Wmt22 | 85.79 | Score (%) | 96.2 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 96.43 | Score (%) | 96 |
| Open Japanese LLM - QA | 71.26 | Score (%) | 95.9 |
| pfgen-bench - Completion Mode - Fluency | 0.85 | Fluency Score | 95.8 |
| Open Japanese LLM - ALT J TO E Comet Wmt22 | 89.13 | Score (%) | 95.3 |
| Open Japanese LLM - AIO Char F1 | 85.92 | Score (%) | 95.2 |
| pfgen-bench - Completion Mode - Score | 0.74 | pfgen Score (mean of three) | 94.9 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-swallow-70b-instruct-v0-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.