Text-to-speech

Fun-CosyVoice3 0.5B (2512) requirements, license & speed

Name: Fun-CosyVoice3 0.5B (2512)
Author: Alibaba FunAudioLLM

By Alibaba FunAudioLLM · released 2025-12-11 · 0.5B parameters

Hugging Face ↗ GitHub ↗ Compare

Commercial useYes

Min memoryNot stated

Runs on CPUUnknown

StreamingYes

Key facts

License: apache-2.0 source
Commercial use: Yes source
Apache-2.0
Parameters: 0.5B source
from model name
Minimum memory: Not stated by source
Runs on CPU: Unknown
Streaming: Yes source
Bi-streaming (text in/audio out), ~150 ms latency claimed
Voice cloning: Yes source
Multilingual / cross-lingual zero-shot
Languages: zh, en, fr, es, ja, ko, it, ru, de source
Released: 2025-12-11 source
Hugging Face repo creation date (proxy for release date)
Gated on HF: No

Will it run on…

Pi 5 Unknown
CPU only Unknown
Apple Silicon Estimate from parameter count
8 GB GPU Estimate from parameter count
12–16 GB GPU Estimate from parameter count
24 GB+ GPU Estimate from parameter count

Pages exist only where we measured it or found a sourced statement. Other rows are inferred from stated memory or parameter count. How we decide

Open the picker →

Affiliate links · disclosed

When self-hosting isn't worth it

Low volume, or you need a managed SLA? A hosted voice API can be cheaper than keeping a machine on. These are affiliate links: we may earn a commission if you sign up. They don't change anything on this site.

ElevenLabs Affiliate
Fish Audio API Affiliate

Measured speed & memory

Benchmark pending. We haven't measured Fun-CosyVoice3 0.5B (2512) yet. Planned targets: Raspberry Pi 5, CPU-only x86, Apple Silicon Mac and consumer GPUs. Get notified.

Accuracy

We don't run accuracy evals. For quality and intelligibility, check the Hugging Face Open TTS Leaderboard.

Similar TTS models

Spotted an error? Report it. Every fact links to the model card or repo it came from.

Key facts

Measured speed & memory

Accuracy

Similar TTS models

New open speech models: will they run on your hardware?