Llama3.3-Swallow-70B: benchmark results
Provider: Meta.
Aggregate Rating 1685 ± 48, rank #420 of 10916 rated models, from 2 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Japanese G2P (Parse Mode) | 2.85 | Kana character error rate (%; can exceed 100) in parse mode | 75 |
| Benchmarking Large Language Models for Graphem | 11.78 | Direct (self-reported) | 69.4 |
Interactive version: theaggregate.ai/model?slug=llama3-3-swallow-70b · How It Works · Data refreshed daily, snapshot 2026-11-08.