AI audio and music models
12 models · from 1 credits
At a glance
- Models
- 12
- Cheapest to try
- Gemini 3.8 Flash TTS · 1 credit
- Newest
- Gemini 3.8 Flash TTS · September 22, 2026
- Makers
- Suno, Google, MiniMax, ByteDance, xAI, Alibaba, Resemble AI
The models
- TopSuno v6Suno · AudioFull songs with vocals from your lyrics or a short briefFrom 72 credits per run
- TopGemini 3.8 Flash TTSGoogle · AudioExpressive speech that acts out cues like [laughing]From 0.32 credits per second of audio
- TopMiniMax Speech 2.8 HDMiniMax · AudioHigh-fidelity speech in 400+ voices with emotion controlFrom 120 credits per 1,000 characters
- MiniMax Music 3.0MiniMax · AudioFinished stereo songs, with vocals or instrumentalFrom 180 credits per run
- Seed Audio 1.0ByteDance · AudioMulti-voice audio scenes with music and effectsFrom 360 credits per run
- NewGemini 3.8 Flash-Lite TTSGoogle · AudioFast, lightweight speech in 30 voicesFrom 0.21 credits per second of audio
- Grok TTSxAI · Audio71 voices across 20 languages with speech tagsFrom 18 credits per 1,000 characters
- Seed Speech 2.0ByteDance · Audio41 voices with tone set in plain languageFrom 72 credits per 1,000 characters
- MiniMax Speech 2.8 TurboMiniMax · AudioFast speech in 400+ voices with emotion controlFrom 72 credits per 1,000 characters
- Qwen3 TTSAlibaba · AudioOpen-source speech with tone and pace set in plain wordsFrom 68 credits per 1,000 characters
- ChatterboxResemble AI · AudioEnglish speech that can match a short voice sampleFrom 36 credits per run
- Gemini 2.5 Pro TTSGoogle · AudioDirectable voices for narration and two-person dialogueFrom 0.45 credits per second of audio
Quick answers
Which is the cheapest to try?
Gemini 3.8 Flash TTS starts at 1 credits for one run.
Which one is the newest?
Gemini 3.8 Flash TTS, released on September 22, 2026.
Which one should I start with?
Start with Suno v6, the most used one here. Full songs with vocals from your lyrics or a short brief