Voxtral 3 TTSvsStep Audio 2

Voxtral 3 TTS leads the overall Podium Score by 1.2 points (90.2 vs 89.0).Category wins: Voxtral 3 TTS 4 — 1 Step Audio 2.Step Audio 2 is the cheaper pick ($0.024/M vs $0.03/M per 1M output tokens). Voxtral 3 TTS is faster (165 t/s vs 150 t/s).

Open in Compare Tool →View Voxtral 3 TTSView Step Audio 2
Voxtral 3 TTSDrag slider or hover to compareStep Audio 2
Step Audio 2
89.0
Coding: 0.0 · Speed: 150 t/s
Output Price: $0.024/M
Voxtral 3 TTS
90.2
Coding: 0.0 · Speed: 165 t/s
Output Price: $0.03/M
Quick VerdictAggregated 2026 Benchmark Analysis

Voxtral 3 TTS is rated higher overall with a Podium Score of 90.2 (vs 89.0 for Step Audio 2). In head-to-head domain evaluations, Voxtral 3 TTS outperforms in 4 of 7 benchmark categories.

Coding & Engineering
Voxtral 3 TTS (0.0)
Inference Speed
Voxtral 3 TTS (165 t/s)
Cost Efficiency
Step Audio 2 ($0.024/M)
API Price / 1M (in · out)
Voxtral 3 TTS $0.015/M · $0.03/M — Step Audio 2 $0.012/M · $0.024/M

Choose Voxtral 3 TTS for vision and multimodal apps, and Step Audio 2 for autonomous agents and tool use.

MetricVoxtral 3 TTSStep Audio 2
Podium Score90.2 ▲89.0
Arena Elo1438 ▲1422
Intelligence Index90.0 ▲89.0
Coding0.00.0
Math0.00.0
Reasoning33.0 ▲32.0
Agentic52.055.0 ▲
Knowledge47.0 ▲46.0
Multimodal87.0 ▲85.0
Long-context48.0 ▲45.0
Output speed165 t/s ▲150 t/s
Output price$0.03/M$0.024/M ▲
Context window32K64K ▲
Scores aggregated from five independent leaderboards. See Methodology. More head-to-heads: Voxtral 3 TTS vs Claude Mythos · Voxtral 3 TTS vs Claude 5.5 Sonnet · Voxtral 3 TTS vs Kimi K3.1 Pro · Voxtral 3 TTS vs Eleven Multilingual v3
0 / 4
Compare Now →