IndexTTS-2.5
Official site ↗
Open weights — free
+ Bilibili's open zero-shot cloning — 5 languages, emotion and duration control
− Check the bilibili model license for commercial use
MODEL TOOL
Voice & TTS：Budget
Voiceover, audio content, voice cloning — ranked against the Artificial Analysis Speech Arena.
1 · What are you doing?
2 · Budget tier
Recommended for · Voice & TTS · Budget
Open-source voices — free local narration drafts and bulk reads.
Official site ↗
CosyVoice 2
Official site ↗
Open weights — free
+ Great dialect coverage and fine control
− Setup work required
Kokoro
Official site ↗
Open weights — Apache 2.0
+ 82M params, runs on CPU
− Limited emotion and voices
Video generationMiniMax H3 Open Weights / Wan2.2（Apache 2.0） / LTX-2.5 Fast / ProImage generationFLUX.2 dev / Z-Image Turbo / Qwen-ImageVoice & TTSIndexTTS-2.5 / CosyVoice 2 / KokoroAI avatarHunyuanVideo-Avatar / EchoMimic3D generationTRELLIS.2 / Hunyuan3D 2.1 / Pixal3DMusic generationMiniMax Music 3.0 / Stable Audio 3 Medium / ACE-Step 1.5
How to read this
Three tiers: flagship for quality-critical work, balanced for daily production, budget for bulk and experiments. Names track the current mainstream options — the landscape moves fast, so treat the tier logic (not the exact model list) as the durable part.
Rankings and prices checked — leaderboards move weekly; always confirm pricing on the official site before committing.
Sources: Artificial Analysis · LMArena · Crafiq 3D · Taxon3D