Open Models Almanac
Methodsmodel architecturemultimodal architecture

general family · filed under model architecture

Native multimodal architecture

Processes multiple modalities as native inputs within one model rather than relying on a separate post-hoc alignment stage; the evidence spans differing modality sets and implementations.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See multimodal architecture for the methods filed beside it.

Also called native multimodal encoding, natively multimodal, Native multimodal input, single shared backbone, Native multimodal understanding, native multimodal in one model.

sources
5
models
7
labs adopt it
5
strongest
core

Documented in

Evidence

5 quotes from 5 sources, one card per model, each opening with its strongest.

How sources treat it
core 5

Kimi K3

Moonshot AIadopted1 source

text, images, and videos are processed by a single shared backbone within one context, with no post-hoc modality-alignment stage.

coremodel architecturein Kimi K3Moonshot AI

Inkling

Thinking Machines Labadopted1 source

The model is natively multimodal — images and video are encoded via a hierarchical patch encoder, and audio via discrete token encoding — with all modalities projected into a shared hidden space and processed jointly by the decoder.

coremodel architecturein InklingThinking Machines Lab

MiMo-V2.5-Pro

Xiaomiadopted1 source

Native multimodal in one model. V2-Pro was text-and-code only. Multimodal capability existed in the separate V2-Omni model... V2.5 collapses both into a single architecture: native image, video, and audio understanding with Pro-level reasoning.

coremodel architecturein MiMo-V2.5-ProXiaomi
↑