vikhyatk / image-text-to-text updated 10 months ago

moondream2

Name: moondream2
Author: vikhyatk

⚠️ This repository contains the latest version of Moondream 2, our previous generation model. The latest version of Moondream is Moondream 3 (Preview).

Params

1.9 B

Context

—

Downloads 30d

2.6 M

Likes

1,429

Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

2.6 M2.3 M

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	3.9 GB	5.0 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture: HfMoondream
Parameters: 1.9 B
Tensor type: BF16
Licence: apache-2.0
First seen on the Hub: 2024-03-04
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

Qwen3.5-9B 12.4 M Qwen2.5-VL-3B-Instruct 8.2 M Qwen3.5-4B 6.7 M GLM-OCR 3.8 M Qwen2-VL-2B-Instruct 3.0 M Qwen3.5-2B 2.6 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB