c4ai-aya-expanse-32B: benchmark results
Provider: Cohere. Released 2024-10-24. Access: Open.
Aggregate Rating 1497 ± 41, rank #2410 of 10916 rated models, from 49 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Enkrypt AI - Safety Risk | 19.55 | Risk Score | 87.4 |
| Enkrypt AI - Insecure Code Risk | 9.78 | Risk Score | 80.4 |
| Enkrypt AI - Jailbreak Risk | 4.67 | Risk Score | 78.1 |
| Enkrypt AI - Toxicity Risk | 1.64 | Risk Score | 77.4 |
| Enkrypt AI - Risk Score | 26.59 | Risk Score | 71.3 |
| BALSAM - Sequence Tagging | 35.29 | Overall score (0-100, LLM-judged generation and multiple cho | 48.3 |
| Enkrypt AI - Bias Risk | 84.75 | Risk Score | 44.3 |
| BALSAM - Fill in the Blank | 71.67 | Overall score (0-100, LLM-judged generation and multiple cho | 43.3 |
| BALSAM - Entailment | 71.08 | Overall score (0-100, LLM-judged generation and multiple cho | 36.7 |
| BALSAM - Text Classification | 35.36 | Overall score (0-100, LLM-judged generation and multiple cho | 36.7 |
| Vectara Hallucination Leaderboard | 89.1 | Factual Consistency Rate (%) | 34.5 |
| Chatbot Arena (Text - Non-English) | 1253 | Arena Score | 26.9 |
Interactive version: theaggregate.ai/model?slug=c4ai-aya-expanse-32b · How It Works · Data refreshed daily, snapshot 2026-11-08.