DeepSeekReleased on Dec 13, 2024

DeepSeek VL2: Benchmarks, Pricing & Context Window

DeepSeek VL2 is a language model from DeepSeek, released in December 2024, with multimodal input.

An advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL. DeepSeek-VL2 demonstrates superior capabilities across various tasks, including but not limited to

Input
TextImage
Output
Text

DeepSeek VL2 API

API access coming soon

DeepSeek VL2 will be available through our gateway shortly.

DeepSeek VL2 examples

Recent arena outputs from DeepSeek VL2, picked from the highest-ranked matchups.

DeepSeek VL2 license

DeepSeek VL2 is released under the deepseek license, which restricts commercial use, has 27.0B parameters.

License
deepseek
Non-commercial
Parameters
27.0B

FAQ

Common questions about DeepSeek VL2.

What is the DeepSeek VL2 release date?

DeepSeek VL2 was released on December 13, 2024 by DeepSeek.

Who created DeepSeek VL2?

DeepSeek VL2 was created by DeepSeek.

How many parameters does DeepSeek VL2 have?

DeepSeek VL2 has 27.0 billion parameters.

What is the license for DeepSeek VL2?

DeepSeek VL2 is released under the deepseek license.

Is DeepSeek VL2 multimodal?

Yes, DeepSeek VL2 is a multimodal model that can process both text and images as input.