MO

moondream2

MoondreamGGUF by moondreamUpdated 2025-05-22
Download on Hugging Face
Parameters1.42B
Context2K
Layers24
Architecturephi2
Smallest quant2.8 GB
Quick answer

moondream2 needs about 5.0 GB of memory at F16 (a 2.8 GB file) with an 8K context window, so it fits on a 6 GB GPU.

Moondream2: 1.42B‑parameter vision‑language model with efficient 4‑bit quantization

Moondream2 is a 1.42‑billion‑parameter phi2 architecture model trained on 2048‑token context. The 4‑bit release runs on 2.84 GB GGUF file and uses about 5 GB of RAM at 8 k context. It is licensed under Apache‑2.0.

Good at

  • Vision‑language tasks
  • Coding assistance
  • Efficient inference on modest GPUs

Limits

  • No vision support beyond image‑to‑text
  • Limited to 2048‑token context
  • Requires 8 k context for full memory estimate

Best for

  • Image captioning
  • Code generation from images
  • Low‑latency inference on consumer GPUs

Who it’s for. Developers needing a lightweight vision‑language model for edge devices or local inference. The model is suitable for applications that require both image understanding and code generation within a single model.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F162.8 GB 3.7 GB
Fits Get file →

The only available quant is F16, 2.84 GB in size. It requires approximately 5 GB of RAM at 8 k context. For 8 GB VRAM setups, run with 8 k context and F16 quantization.

What GPU do you need for moondream2?

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: the best quant that fits is F16 (2.8 GB file, about 5.0 GB in use).

Which moondream2 quant should you choose?

F16 (2.8 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturephi2
Parameters1.42B
Layers24
Trained context2,048 tokens
KV cache (fp16)192 KB per token

Source

Quantized bymoondream
FamilyMoondream
Licenseapache-2.0
Downloads (HF)6,889
Last updated2025-05-22

Frequently asked questions about moondream2

How much VRAM does moondream2 need?

About 5.0 GB at F16 with an 8K context window. That is the 2.8 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 moondream2 on an 8 GB GPU?

Yes, with F16: the 2.8 GB file needs about 5.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of moondream2 GGUF?

It depends on the quant: F16 is 2.8 GB.

What context length does moondream2 support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.4 GB at 2K.

Where can I download moondream2 GGUF?

The files are on Hugging Face in the moondream/moondream-2b-2025-04-14-4bit repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does moondream2 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 Moondream2 on an 8 GB VRAM GPU?

Yes, the 2.84 GB GGUF file fits in 8 GB VRAM. Expect about 5 GB of RAM usage at 8 k context.

Is Moondream2 open source?

Yes, it is released under the Apache‑2.0 license.

Similar models

Browse all →

More from moondream

All by moondream →

More Moondream models

All Moondream →

Other tiny models

All tiny models →