Skip to content
@Eximius-Labs

Eximius Labs

Eximius Labs

One vector space for text, image, video, audio, and sensors — plus a perception layer. Open weights, self-hostable.


Eximius Labs builds open-weight multimodal models that run entirely on your own hardware. The stack has two complementary families.

Fusion Embedding is the memory layer. It maps text, images, video, audio, and sensor streams into a single shared vector space for retrieval, RAG, clustering, and cross-modal search. The weights are published and the models run on your own hardware.

Fusion Perception is the perception layer. Dense scene understanding and geometric place recognition on a frozen vision backbone, with a projector that drops its features into the Fusion Embedding space.

The idea

We extend a state-of-the-art frozen base with new modalities and senses without modifying a single base weight. Text, image, and video vectors stay bit-for-bit identical to the base model, so gaining a modality never costs you a re-index.

Fusion Embedding — the memory layer

Model Description
fusion-embedding-2 Current line. Connector plus modality-gated deep adapters, 60.6M trained parameters.
fusion-embedding-1 Connector-only architecture, 16.4M trained parameters. Final at v0.3.
Ember Thermal sense pack: a separately loadable adapter pack that adds infrared to fusion-embedding-2 as a fifth modality, 44.2M trained parameters.
Tremor Inertial sense pack: reads a robot's or wearable's accelerometer as language. Ships a general base and a Unitree-G1 head.

Sense packs are named for the physical trace their sensor reads. Ember reads heat; Tremor reads motion. More senses are in training.

Fusion Perception — the perception layer

Model Description
fusion-perception-1 Frozen DINO backbone plus light heads: geometric place recognition ("where am I"), segmentation, and a projector that makes those features language-searchable in the Fusion Embedding space.

Start here

Popular repositories Loading

  1. fusion-embedding fusion-embedding Public

    One embedding space for text, image, video, and audio. Open weights, self-hostable.

    Python 6

  2. .github .github Public

    Organization profile

  3. fusion-perception fusion-perception Public

    Fusion Perception - the perceive model of the Eximius stack (frozen DINO backbone + task heads; place recognition, segmentation, and a projector into the Fusion Embedding shared space).

    Python

Repositories

Showing 3 of 3 repositories

Top languages

Loading…

Most used topics

Loading…