Orpheus 3B
A solid pick for self-hosting and full data control.
Overview
Orpheus 3B is an open-weight text-to-speech model released by Canopy Labs under an Apache 2.0 licence. At 3 billion parameters it's small enough to run on laptop-class hardware, making it a natural fit for on-premise setups where generated audio stays entirely on the organisation's own infrastructure, regardless of the vendor's home country. With no published model card, actual audio quality and language coverage remain unverified, and the commercial terms implied by the licence still need legal confirmation before production use.
Skill profile
Not disclosed
Strengths
- Open-source and self-hostable
Limitations
- API pricing not disclosed
Who is it for
- you want to control cost or self-host
- you have GDPR constraints
Ideal use cases
- Local voice prototyping
- Embedded voice assistants
- Internal content dubbing
- Laptop-based TTS testing
Access & availability
Key specifications
Privacy
Run it locally
| Quantization | Disk | RAM / VRAM | Typical hardware |
|---|---|---|---|
| Q4 · recommended | 1.7 GB | 4 GB | Any recent PC/Mac |
| Q8 · balanced | 3.2 GB | 6 GB | Any recent PC/Mac |
| FP16 · max quality | 6 GB | 9 GB | 16GB PC / M1+ Mac / 8GB GPU |
Estimates for a moderate context. Long contexts need more RAM (KV cache).
Deploy
Copy-ready commands generated from this card. Adjust context length and GPU count to your hardware.
OpenAI-compatible server for production on NVIDIA GPUs.
pip install vllm
vllm serve canopylabs/orpheus-3b-0.1-ft \
--max-model-len 32768 \
--tensor-parallel-size 1 \
--dtype autoAdvanced data · for expertsArchitecture, modalities, detailed cost, full benchmarks▾
| Arena Elo | — |
| MMLU | — |
| GPQA | — |
| HumanEval | — |
| SWE-Bench | — |
| MATH | — |