Google Releases Compact Gemma 4 E2B Multimodal Model
The new 2-billion-parameter model from Google DeepMind brings efficient image-and-text understanding to the open-source Gemma family.
Google DeepMind has expanded its open-source offerings with the release of Gemma 4 E2B, a new model in the Gemma 4 family. This compact 2-billion-parameter model is designed for efficiency and introduces multimodal capabilities, allowing it to process both images and text to generate text-based responses.
Unlike previous text-only Gemma models, Gemma 4 E2B is a vision-language model (VLM). This means developers can provide it with an image alongside a text prompt to perform tasks like visual question answering, image description, and other forms of visual reasoning. The "E2B" in its name likely signifies its focus on being an "Efficient 2 Billion" parameter model.
Why It Matters
The release of a small, capable VLM like Gemma 4 E2B is significant for making advanced AI more accessible. Its modest size makes it suitable for running on consumer hardware, in edge devices, or in cloud environments where computational cost is a concern. This democratizes access to multimodal technology that has often been restricted to much larger, more demanding models.
The model is available under the Gemma license, and developers can explore its architecture and weights on its official Hugging Face repository.
Sources
- Visit
google/gemma-4-E2B
Hugging Face
More from Google DeepMind
All Google DeepMind releases →
EmbeddingGemma 2 Expands to Any Modality
Google DeepMind's compact embedding model now spans text, vision, audio, and video in a single vector space.
Google DeepMind's Gemma 4 Goes Multimodal and MoE
The new open-weights family adds a mixture-of-experts design, encoder-free multimodal inputs, and an optional thinking mode.
Google DeepMind Releases TabFM for Tabular Data
A new foundation model brings zero-shot, in-context learning to classification and regression on structured tables.
More in Any-to-Any
All Any-to-Any →
Liquid AI's LFM3-Omni-2B brings any-to-any AI to the edge
The 2-billion-parameter model combines vision, audio and speech-to-speech in a package small enough to run on device.
MiniMax Debuts M3.1 Flash, a Multimodal MoE Model
The new release pairs a mixture-of-experts design with agentic, coding, and video-understanding abilities in a single open model.
IBM's Granite Switch 4.1 Unifies Chat, Logic, Speech
A single endpoint now routes text conversation, decision-making, and speech synthesis across IBM's Apache-licensed Granite line.
0 comments
No comments yet. Be the first to weigh in.