# Zonos 2 model evaluation: Voice Benchmark Profile

> Zyphra’s ZONOS2 GGUF model card reports word error rate, speaker similarity, and UTMOS for the reference F16 build.

- Scope: WER, speaker similarity, and UTMOS naturalness for the F16 reference build on Zyphra’s evaluation set
- Measurement lanes: Voice experience
- Primary metric: WER (lower is better), speaker similarity and UTMOS (higher is better)
- Available evidence: Exact figures published on the official GGUF model card
- Owner: Zyphra
- Source snapshots refreshed: 2026-09-11

## Interpretation limit

Provider-run evaluation on Zyphra’s own set; the blog compares against other systems qualitatively only.

## Primary sources

- [Owner page](https://huggingface.co/Zyphra/ZONOS2-GGUF)

Canonical page: https://benchlm.ai/voice-benchmarks/zonos-2-launch-evaluation