VO

VoxCPM-0.5B

VoxcpmGGUF by bluryarUpdated 2026-04-21
Download on Hugging Face
Parameters728M
Context32K
Layers24
Architecturellama
Smallest quant0.5 GB
Quick answer

VoxCPM-0.5B needs about 2.1 GB of memory at Q4_K (a 1.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 21 quants, from 0.5 GB (Q4_K) to 9.5 GB (F32). Longer context needs more: at 32K tokens the KV cache alone adds about 0.4 GB.

VoxCPM‑0.5B: Tokenizer‑free TTS model for expressive speech and voice cloning

VoxCPM‑0.5B is a 0.728 billion‑parameter Llama‑based model designed for continuous‑space speech synthesis. It supports context‑aware speech generation and zero‑shot voice cloning from short audio clips.

Good at

  • Context‑aware expressive speech
  • Zero‑shot voice cloning
  • Streaming synthesis with low RTF

Limits

  • No vision support
  • License not specified
  • Requires 8 k context for memory estimates

Best for

  • Real‑time TTS applications
  • Voice cloning demos
  • Low‑latency streaming synthesis

Who it’s for. Developers building TTS or voice cloning tools who need a lightweight, continuous‑space model. It is suitable for research and production environments that can run llama.cpp inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K0.5 GB 1.1 GB
Fits Get file →
Q4_K0.6 GB 1.2 GB
Fits Get file →
Q4_K0.6 GB 1.2 GB
Fits Get file →
Q4_K0.7 GB 1.3 GB
Fits Get file →
Q8_00.8 GB 1.4 GB
Fits Get file →
Q8_00.9 GB 1.5 GB
Fits Get file →
Q8_01.0 GB 1.6 GB
Fits Get file →
Q8_01.0 GB 1.6 GB
Fits Get file →
F161.5 GB 2.1 GB
Fits Get file →
F161.5 GB 2.1 GB
Fits Get file →
Q4_K1.5 GB 2.1 GB
Fits Get file →
Q4_K1.6 GB 2.2 GB
Fits Get file →
F161.8 GB 2.4 GB
Fits Get file →
F161.8 GB 2.4 GB
Fits Get file →
Q8_02.6 GB 3.2 GB
Fits Get file →
Q8_02.6 GB 3.2 GB
Fits Get file →
F322.9 GB 3.5 GB
Fits Get file →
F323.6 GB 4.2 GB
Fits Get file →
F164.8 GB 5.4 GB
Fits Get file →
F164.8 GB 5.4 GB
Fits Get file →
F329.5 GB 10.1 GB
Too big Get file →

For 8 GB VRAM use Q4_K (file size 0.5 GB, ~1.1 GB memory). For 12 GB VRAM use Q8_0 (file size 0.8 GB, ~1.4 GB memory). For 16 GB VRAM use F16 (file size 1.46 GB, ~2.1 GB memory).

What GPU do you need for VoxCPM-0.5B?

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 F32 (2.9 GB file, about 3.5 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (4.8 GB file, about 5.4 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (4.8 GB file, about 5.4 GB in use).
  • 12 GB VRAM: the best quant that fits is F32 (9.5 GB file, about 10.1 GB in use).

Which VoxCPM-0.5B quant should you choose?

Q4_K (1.5 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

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

Architecturellama
Parameters728M
Layers24
Trained context32,768 tokens
KV cache (fp16)12 KB per token

Source

Quantized bybluryar
FamilyVoxcpm
Downloads (HF)2,493
Last updated2026-04-21

Frequently asked questions about VoxCPM-0.5B

How much VRAM does VoxCPM-0.5B need?

About 2.1 GB at Q4_K with an 8K context window. That is the 1.5 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 VoxCPM-0.5B on an 8 GB GPU?

Yes, with F16: the 4.8 GB file needs about 5.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of VoxCPM-0.5B GGUF?

It depends on the quant: Q4_K is 0.5 GB, Q4_K is 0.6 GB, Q4_K is 0.6 GB, Q4_K is 0.7 GB, Q8_0 is 0.8 GB, Q8_0 is 0.9 GB and 15 more.

What context length does VoxCPM-0.5B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.4 GB at 32K.

Where can I download VoxCPM-0.5B GGUF?

The files are on Hugging Face in the bluryar/VoxCPM-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run VoxCPM‑0.5B on 8 GB VRAM?

Yes, the Q4_K quantization requires about 1.1 GB of VRAM at 8 k context, which fits within 8 GB.

Does VoxCPM‑0.5B support image input?

No, the model has no vision capabilities and is focused solely on audio generation.

What is the maximum context length for this model?

The trained context length is 32768 tokens, but memory estimates are given for 8 k context.

Other GGUFs of VoxCPM-0.5B

Similar models

Browse all →

More from bluryar

All by bluryar →

More Voxcpm models

All Voxcpm →

Other tiny models

All tiny models →