orpheus-finetuned-3b needs about 3.9 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 42 quants, from 0.0 GB to 7.6 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 3.8 GB.
Orpheus 3B finetuned for text-to-speech
Orpheus 3B finetuned is a 3.78B parameter Llama-based TTS model. It supports human-like speech synthesis with zero‑shot voice cloning and guided emotion tags. The model is released under Apache‑2.0 and can be run with llama.cpp.
Good at
- Human‑like TTS
- Zero‑shot voice cloning
- Emotion control
Limits
- No vision support
- Large context requires 131072 tokens
- Requires 8‑bit or higher quant for full performance
Best for
- Real‑time TTS applications
- Voice cloning demos
- Emotion‑tagged speech synthesis
Who it’s for. Developers building TTS systems or voice assistants who need high‑quality, expressive speech. The model is suitable for research and production use under an Apache‑2.0 license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 1.4 GB | Fits | Get file → | |
| IQ2_XXS | 1.5 GB | 2.9 GB | Fits | Get file → | |
| IQ2_XS | 1.6 GB | 3.0 GB | Fits | Get file → | |
| IQ2_S | 1.6 GB | 3.1 GB | Fits | Get file → | |
| IQ2_M | 1.7 GB | 3.1 GB | Fits | Get file → | |
| Q2_K_S | 1.7 GB | 3.1 GB | Fits | Get file → | |
| IQ3_XXS | 1.8 GB | 3.2 GB | Fits | Get file → | |
| Q2_K_M | 1.8 GB | 3.3 GB | Fits | Get file → | |
| IQ3_XS | 1.8 GB | 3.3 GB | Fits | Get file → | |
| IQ3_M | 2.0 GB | 3.4 GB | Fits | Get file → | |
| IQ3_S | 2.0 GB | 3.4 GB | Fits | Get file → | |
| Q3_K_S | 2.0 GB | 3.5 GB | Fits | Get file → | |
| Q2_K_L | 2.1 GB | 3.5 GB | Fits | Get file → | |
| Q4_0 | 2.1 GB | 3.6 GB | Fits | Get file → | |
| Q3_K_M | 2.1 GB | 3.6 GB | Fits | Get file → | |
| IQ4_XS | 2.2 GB | 3.6 GB | Fits | Get file → | |
| IQ4_NL | 2.3 GB | 3.7 GB | Fits | Get file → | |
| Q4_1 | 2.4 GB | 3.8 GB | Fits | Get file → | |
| Q3_K_L | 2.4 GB | 3.8 GB | Fits | Get file → | |
| Q4_K_S | 2.4 GB | 3.9 GB | Fits | Get file → | |
| Q4_K_M | 2.5 GB | 3.9 GB | Fits | Get file → | |
| Q5_0 | 2.6 GB | 4.0 GB | Fits | Get file → | |
| Q4_0 | 2.6 GB | 4.1 GB | Fits | Get file → | |
| Q4_K_L | 2.7 GB | 4.1 GB | Fits | Get file → | |
| Q5_K_S | 2.8 GB | 4.2 GB | Fits | Get file → | |
| Q4_1 | 2.8 GB | 4.2 GB | Fits | Get file → | |
| Q5_K_M | 2.8 GB | 4.2 GB | Fits | Get file → | |
| Q5_1 | 2.8 GB | 4.3 GB | Fits | Get file → | |
| Q5_0 | 3.0 GB | 4.4 GB | Fits | Get file → | |
| Q5_K_L | 3.0 GB | 4.5 GB | Fits | Get file → | |
| Q6_K_M | 3.1 GB | 4.6 GB | Fits | Get file → | |
| Q5_1 | 3.1 GB | 4.6 GB | Fits | Get file → | |
| Q6_K_L | 3.3 GB | 4.8 GB | Fits | Get file → | |
| Q8_0 | 4.0 GB | 5.5 GB | Fits | Get file → | |
| BF16 | 5.0 GB | 6.5 GB | Fits | Get file → | |
| F16 | 5.0 GB | 6.5 GB | Fits | Get file → | |
| BF16 | 5.5 GB | 6.9 GB | Fits | Get file → | |
| F16 | 5.5 GB | 6.9 GB | Fits | Get file → | |
| BF16 | 5.9 GB | 7.4 GB | Tight | Get file → | |
| F16 | 5.9 GB | 7.4 GB | Tight | Get file → | |
| BF16 | 7.6 GB | 9.0 GB | Too big | Get file → | |
| F16 | 7.6 GB | 9.0 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (file size 2.14 GB, memory 3.6 GB). For 12 GB VRAM, Q5_0 (file size 2.97 GB, memory 4.4 GB) offers higher precision. For 16 GB VRAM, Q6_K_L (file size 3.35 GB, memory 4.8 GB) gives the best trade‑off between size and quality.
What GPU do you need for orpheus-finetuned-3b?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q5_0 (2.6 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.0 GB file, about 5.5 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (5.9 GB file, about 7.4 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (7.6 GB file, about 9.0 GB in use).
Which orpheus-finetuned-3b quant should you choose?
Q4_K_M is the usual starting point for orpheus-finetuned-3b: it balances file size and quality well. Here it is a 2.5 GB file.
The Q8_0 file is 4.0 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about orpheus-finetuned-3b
How much VRAM does orpheus-finetuned-3b need?
About 3.9 GB at Q4_K_M with an 8K context window. That is the 2.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run orpheus-finetuned-3b on an 8 GB GPU?
Yes, with BF16: the 5.9 GB file needs about 7.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of orpheus-finetuned-3b GGUF?
It depends on the quant: file is 0.0 GB, IQ2_XXS is 1.5 GB, IQ2_XS is 1.6 GB, IQ2_S is 1.6 GB, IQ2_M is 1.7 GB, Q2_K_S is 1.7 GB and 36 more.
What context length does orpheus-finetuned-3b support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.9 GB at 8K and 3.8 GB at 32K.
Where can I download orpheus-finetuned-3b GGUF?
The files are on Hugging Face in the Mungert/orpheus-finetuned-3b-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does orpheus-finetuned-3b use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Orpheus 3B on 8 GB VRAM?
Yes, using Q4_0 quantization requires about 3.6 GB VRAM at 8 k context.
Does the model support vision?
No, the model has no vision capabilities.
What license applies to this model?
The model is released under the Apache‑2.0 license.