model

VoxCPM2

Tokenizer-free text-to-speech model from OpenBMB that generates continuous speech representations with a diffusion-autoregressive architecture, supporting voice cloning and text-prompted voice design in one checkpoint.

Known aliases

openbmb/VoxCPM2
VoxCPM

Relationships

No evidence-backed relationships are recorded.

Current clusters

build1 publisher

VoxCPM2 trades the per-character speech bill for a GPU and a base-URL change

OpenBMB's 2B-parameter model is Apache-2.0, speaks 30 languages, and serves through vLLM's OpenAI-compatible /v1/audio/speech. The real-time factor quoted for it was measured on a different backend than the one the project recommends for production.

Publishers:dev.to

Reality

Evidence38
Adoption: Insufficient
Hype gap+38
Incentives42
Confidence40