How to run Janus-Pro-7B locally

DeepSeek · Unified understanding + image gen · 7B params · DeepSeek Model License (code MIT) (commercial OK)

Unified MLLM that decouples vision encoding for understanding versus generation, so one 7B model both answers questions about images and produces images from text. Weights are under the DeepSeek Model License, which permits commercial use.

What it costs to run — $0 markup

On your own machine — $0. Runs free if you have about 16 GB of VRAM (~14GB in BF16, runs on a 16GB card; image generation is at 384x384 so memory is modest. 8-bit fits under 12GB.), via Transformers (Janus repo).
Rent a GPU — from $0.26/hr. Fits RTX 3090 24GB at the direct vendor price ($0 markup) — pay only for the minutes you generate.
Download the weights — free. Open weights at deepseek-ai/Janus-Pro-7B.

Note: generative-media models are billed per image / per second / per minute on hosted services — not per token. Running locally or on your own rented GPU is usually far cheaper and keeps your data on your machine.

Key facts

Does	Text → image, Visual understanding, Image captioning, Visual question answering
VRAM to run	~16 GB (~14GB in BF16, runs on a 16GB card; image generation is at 384x384 so memory is modest. 8-bit fits under 12GB.)
Download	~14 GB
Parameters	7B
License	DeepSeek Model License (code MIT) (commercial use OK)
Run with	Transformers (Janus repo)

Get Janus-Pro-7B on Hugging Face →

More multimodal / omni models

Qwen2.5-Omni-7B — Omni understanding + speech generation, ~24 GB VRAM
OmniGen2 — Any to image (gen + edit + in context), ~17 GB VRAM
MiniCPM-o 2.6 — Omni understanding + speech generation, ~9.0 GB VRAM
BAGEL-7B-MoT — Unified understanding + image gen + edit, ~24 GB VRAM
Emu3.5 — Any to any world model (gen + edit), ~48 GB VRAM
Emu3-Gen — Next token any to any generation, ~18 GB VRAM

Browse: all media models · chat / LLM models