What GPU do I need to run moondream/moondream3-preview?

9.3B parameters, published in BF16. View on Hugging Face

9.3B
Parameters
BF16
Native precision
HfMoondream
Architecture
image-text-to-text
Pipeline

moondream3-preview is published by moondream on Hugging Face, with 166,938 downloads and 682 likes to date. It's a HfMoondream model built for image-text-to-text, published natively in BF16.

VRAM required & cheapest live GPU fit

Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.

PrecisionWeight sizeRequired VRAMCheapest live fitGPUs neededEst. $/hr (full fit)
BF1617.3 GB20.7 GBRTX A50001$0.176/hr
FP8 (quantized)8.6 GB10.4 GBRTX 5060 Ti1$0.180/hr
INT4 (quantized)4.3 GB5.2 GBRTX 30701$0.088/hr

A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.

INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.

Cheapest way to run moondream3-preview at its published (BF16) precision: 1× RTX A5000, at $0.176/hr per GPU ($0.176/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.

moondream3-preview: KV cache by context length

The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.

KV-cache size for moondream3-preview is not published for this architecture, so no figure is shown.

moondream3-preview: common questions

Does moondream3-preview fit on a 24 GB GPU?

Yes. At BF16 it needs 20.7 GB of VRAM, so a 24 GB card holds it with 3.3 GB to spare. A 16 GB card is not enough for it at BF16.

What is the least VRAM moondream3-preview can run in?

5.2 GB, at INT4 (quantized), which fits a 6 GB card, against 20.7 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.

Does quantizing moondream3-preview lower the GPU bill?

Yes. At BF16 the cheapest live fit is one RTX A5000 at $0.176/hr. At INT4 (quantized) it drops to one RTX 3070 at $0.088/hr, provided a quantized checkpoint exists for it.

Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.

More Moondream models

All 3 Moondream models: VRAM and GPU requirements

Alternatives at this size

Other models for image-text-to-text within about a third of moondream3-preview's 9.3B parameters, from other model lines.

More on moondream3-preview

Related reading: RTX A5000 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.

Your own GPU cloud accounts

One console for the GPU clouds you already rent from. Your keys stay on your machine.

$20 a month, flat

Connect your accounts

One console for the GPU clouds you already use.

Connect your accounts for $20 a month, flat. Your provider bills the GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.