Janus-Pro-1B

Parameters

1B

Published

Jan 27, 2025

Janus-Pro-1B is an AI model developed by DeepSeek (China), first published in January 2025. It works in the image generation, vision, language and multimodal domain, on tasks such as image generation, text-to-image and visual question answering.

Epoch AI has no training-compute estimate for this model. The model has 1,000,000,000 parameters. It was trained on roughly 817.9B datapoints. Training ran on 8 NVIDIA A100 SXM4 40 GB for about 216 hours.

Access: Open weights (restricted use). Its weights are openly available. It is built on top of SigLIP 400M,DeepSeek-LLM-1.3b-base. Epoch AI rates the confidence of this record as confident.

Full record

Organization: DeepSeek
Country of organization: China
Domain: Image generation, Vision, Language, Multimodal
Task: Image generation, Text-to-image, Visual question answering
Compute estimation method: Hardware, Operation counting
Parameters: 1,000,000,000
Dataset size: 817.9B
Training hardware: NVIDIA A100 SXM4 40 GB
Chips used: 8
Training time: 216 h
Training power draw: 6.3 kW
Model accessibility: Open weights (restricted use)
Open weights: Yes
Base model: SigLIP 400M, DeepSeek-LLM-1.3b-base
Epoch confidence: Confident