MA

magpie tts.cpp

MagpieGGUF by mudlerUpdated 2026-07-24
Download on Hugging Face
Parameters347M
Context—
Layers—
Architecturemagpie-tts
Smallest quant0.5 GB
Quick answer

magpie tts.cpp needs about 1.2 GB of memory at Q6_K (a 0.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.5 GB (Q4_K) to 1.3 GB (F32).

Magpie TTS Multilingual 357M in GGUF format

Magpie TTS Multilingual 357M is a 0.347 billion‑parameter text‑to‑speech model. It supports 12 languages and five expressive voices. The GGUF build bundles the model, NanoCodec decoder, tokenizer and G2P dictionaries for CPU inference.

Good at

  • 12‑language TTS
  • 5 expressive voices
  • CPU‑friendly inference

Limits

  • No GPU required but 8 k context needs ~1.1 GB VRAM
  • No vision or image support
  • License is NVIDIA Open Model License

Best for

  • Local TTS on laptops
  • Embedded devices with limited GPU
  • Batch synthesis of short utterances

Who it’s for. Developers and hobbyists who need a lightweight, multilingual TTS engine for local deployment. The model runs on any CPU without CUDA or Python dependencies.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K0.5 GB 1.1 GB
Fits Get file →
Q5_K0.6 GB 1.1 GB
Fits Get file →
Q6_K0.6 GB 1.2 GB
Fits Get file →
Q8_00.6 GB 1.2 GB
Fits Get file →
F160.8 GB 1.4 GB
Fits Get file →
F321.3 GB 1.9 GB
Fits Get file →

For 8 GB VRAM, Q4_K or Q5_K (0.54 GB or 0.56 GB) are suitable, using ~1.1 GB RAM at 8 k context. If more memory is available, Q8_0 or F16 provide higher precision. F32 requires 1.29 GB and ~1.9 GB RAM.

What GPU do you need for magpie tts.cpp?

These figures add the file size, an fp16 KV cache at an 8K context window 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 F32 (1.3 GB file, about 1.9 GB in use).

Which magpie tts.cpp quant should you choose?

Q6_K (0.6 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

If memory is tight, the smallest file is Q4_K at 0.5 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemagpie-tts
Parameters347M

Source

Quantized bymudler
FamilyMagpie
Licenseother
Downloads (HF)2,590
Last updated2026-07-24

Frequently asked questions about magpie tts.cpp

How much VRAM does magpie tts.cpp need?

About 1.2 GB at Q6_K with an 8K context window. That is the 0.6 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 magpie tts.cpp on an 8 GB GPU?

Yes, with F32: the 1.3 GB file needs about 1.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of magpie tts.cpp GGUF?

It depends on the quant: Q4_K is 0.5 GB, Q5_K is 0.6 GB, Q6_K is 0.6 GB, Q8_0 is 0.6 GB, F16 is 0.8 GB, F32 is 1.3 GB.

Where can I download magpie tts.cpp GGUF?

The files are on Hugging Face in the mudler/magpie-tts.cpp-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 magpie tts.cpp use?

Hugging Face lists the license as other. 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 Magpie TTS on 8 GB VRAM?

Yes. Q4_K or Q5_K quantizations fit within 8 GB VRAM and use about 1.1 GB RAM at 8 k context.

Does this model support voice cloning?

No. Voice cloning was removed for security reasons.

What languages are supported in the GGUF build?

The GGUF build supports en, es, de, fr, it, pt-BR, hi, vi, ko, ar variants; zh and ja are not yet supported by the C++ tokenizer.

Other GGUFs of magpie_tts_multilingual_357m

Similar models

Browse all →

More from mudler

All by mudler →

More Magpie models

All Magpie →

Other tiny models

All tiny models →