HE

Heron NVILA Lite 15B

HeronGGUF by Kyure-AUpdated 2025-05-14
Download on Hugging Face
Parameters14.8B
Context32K
Layers48
Architectureqwen2
Smallest quant9.0 GB
Quick answer

Heron NVILA Lite 15B needs about 14.2 GB of memory at Q6_K (a 12.1 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 4 quants, from 9.0 GB (Q4_K) to 29.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.

Heron NVILA Lite 15B is a 14.8B-parameter Heron model in GGUF format, quantized by Kyure-A. Its trained context length is 32K tokens. The repository has 4 quant files, from 9.0 GB (Q4_K) to 29.5 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K9.0 GB 11.1 GB
Too big Get file →
Q6_K12.1 GB 14.2 GB
Too big Get file →
Q8_015.7 GB 17.8 GB
Too big Get file →
F1629.5 GB 31.7 GB
Too big Get file →

What GPU do you need for Heron NVILA Lite 15B?

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: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: the best quant that fits is Q4_K (9.0 GB file, about 11.1 GB in use).
  • 16 GB VRAM: the best quant that fits is Q6_K (12.1 GB file, about 14.2 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (15.7 GB file, about 17.8 GB in use).
  • 32 GB VRAM: the best quant that fits is F16 (29.5 GB file, about 31.7 GB in use).

Which Heron NVILA Lite 15B quant should you choose?

Q6_K (12.1 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 9.0 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

Architectureqwen2
Parameters14.8B
Layers48
Trained context32,768 tokens
KV cache (fp16)192 KB per token

Source

Quantized byKyure-A
FamilyHeron
Licenseapache-2.0
Downloads (HF)20
Last updated2025-05-14

Frequently asked questions about Heron NVILA Lite 15B

How much VRAM does Heron NVILA Lite 15B need?

About 14.2 GB at Q6_K with an 8K context window. That is the 12.1 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 Heron NVILA Lite 15B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Heron NVILA Lite 15B GGUF?

It depends on the quant: Q4_K is 9.0 GB, Q6_K is 12.1 GB, Q8_0 is 15.7 GB, F16 is 29.5 GB.

What context length does Heron NVILA Lite 15B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.

Where can I download Heron NVILA Lite 15B GGUF?

The files are on Hugging Face in the Kyure-A/Heron-NVILA-Lite-15B-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 Heron NVILA Lite 15B 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.

Similar models

Browse all →

More from Kyure-A

All by Kyure-A →

More Heron models

All Heron →

Other medium models

All medium models →