RakutenAI-7B-instruct: benchmark results
Provider: Other. Access: Open.
Unified ELO 1481 ± 50, rank #1274 of 2656 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| pfgen-bench - Completion Mode - Fluency | 0.6 | Fluency Score | 56.9 |
| pfgen-bench - Completion Mode - Score | 0.49 | pfgen Score (mean of three) | 55 |
| pfgen-bench - Completion Mode - Helpfulness | 0.13 | Helpfulness Score | 53 |
| pfgen-bench - Completion Mode - Truthfulness | 0.75 | Truthfulness Score | 52.7 |
| pfgen-bench - QA Mode - Truthfulness | 0.73 | Truthfulness Score | 45.8 |
| pfgen-bench - QA Mode - Score | 0.45 | pfgen Score (mean of three) | 40.6 |
| pfgen-bench - QA Mode - Helpfulness | 0.11 | Helpfulness Score | 38.2 |
| pfgen-bench - QA Mode - Fluency | 0.53 | Fluency Score | 37.6 |
| Enkrypt AI - Bias Risk | 86.82 | Risk Score | 32 |
| Enkrypt AI - Insecure Code Risk | 64.44 | Risk Score | 15.5 |
| Enkrypt AI - Jailbreak Risk | 19.73 | Risk Score | 15.3 |
| Enkrypt AI - Risk Score | 48.74 | Risk Score | 9.8 |
Interactive version: theaggregate.ai/model?slug=rakutenai-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-19.