OpenBMB · Apache 2.0
VoxCPM2
A 2B voice model for 30 languages with high-quality 48 kHz audio, voice design and cloning.
How much memory it needs
Memory to run each model, from its model card or the size of its files. Smaller versions trade some quality for less memory.
| Model | Version | Needs | Business use |
|---|---|---|---|
| VoxCPM2 | BF16 | 8.0 GB | Allowed |
Which computers run it
The best model in this group for each machine.
| Computer | Best version | Fit |
|---|---|---|
| Laptop or PC, no graphics card, 32 GB | VoxCPM2 BF16 | Needs a graphics card |
| RTX 4060 · 8 GB | VoxCPM2 BF16 | Spills into system RAM |
| RTX 3060 · 12 GB | VoxCPM2 BF16 | Runs well |
| RTX 5060 Ti · 16 GB | VoxCPM2 BF16 | Runs great |
| RTX 4090 · 24 GB | VoxCPM2 BF16 | Runs great |
| RTX 5090 · 32 GB | VoxCPM2 BF16 | Runs great |
| Mac with M4 · 16 GB | VoxCPM2 BF16 | Runs well |
| Mac with M4 Pro · 48 GB | VoxCPM2 BF16 | Runs great |
| Mac with M5 Max · 128 GB | VoxCPM2 BF16 | Runs great |
| Ryzen AI Max+ 395 · 128 GB | VoxCPM2 BF16 | Runs great |
| DGX Spark · 128 GB | VoxCPM2 BF16 | Runs great |
| RTX PRO 6000 · 96 GB | VoxCPM2 BF16 | Runs great |
| Mac Studio with M5 Ultra · 256 GB | VoxCPM2 BF16 | Runs great |
| Mac Studio with M3 Ultra · 512 GB | VoxCPM2 BF16 | Runs great |
What it is good and bad at
Good at
- High audio quality
- Voice design
Not good at
- About 8 GB of graphics memory
Start it
- VoxCPM2: runs in voxcpm (PyTorch) (also Nano-vLLM, vLLM-Omni, llama.cpp-omni (GGUF), ONNX). Model card