Fun-Audio-Chat-8B: benchmark results
Provider: Other.
Aggregate Rating 1556 ± 49, rank #1492 of 10516 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DuplexSpeechBench-IFEval (IAS) | 47.1 | Instruction Adherence Score (%; deterministic real-time floo | 100 |
| DuplexSpeechBench-IFEval (PAS) | 65.9 | Persona Adherence Score (0-100; GPT-4o-judged persona-consis | 75 |
| PolySpeech-100 | 52.88 | Accuracy (%; spoken four-option reading comprehension from B | 71.4 |
| PolySpeech-100 - Chinese Dialects | 77.06 | Accuracy (%; spoken four-option reading comprehension from B | 71.4 |
| PolySpeech-100 - High-Resource Languages | 84.82 | Accuracy (%; spoken four-option reading comprehension from B | 71.4 |
| SpeechEQ | 36.5 | Trajectory accuracy (x 100; both forced choices of a dialogu | 71.4 |
| LongShOTBench - Core Perception | 25.25 | Weighted Rubric Score (%) | 70.3 |
| LongShOTBench - Information | 30.28 | Weighted Rubric Score (%) | 70.3 |
| LongShOTBench - Medium Videos | 29.67 | Weighted Rubric Score (%) | 70.3 |
| LongShOTBench - Speech | 30.12 | Weighted Rubric Score (%) | 70.3 |
| LongShOTBench - Visual | 29.58 | Weighted Rubric Score (%) | 70.3 |
| LongShOTBench - Audio | 27.16 | Weighted Rubric Score (%) | 68.3 |
Interactive version: theaggregate.ai/model?slug=fun-audio-chat-8b · How It Works · Data refreshed daily, snapshot 2026-11-21.