[killbot5000]

The foundation models themselves will be cheap to deploy, but we’ll still need general purpose inferencing hardware to work along side them, converting latent intermediate layers to useful, application-specific concerns. This may level off the demand for “gpu/tpu” hardware, though, by letting the biggest and most expensive layers move to silicon.

[tim-tday]

How specifically would that work? I’ve seen no framework for that happening.

[killbot5000]

The output of the transformation layers are a collection of embeddings in the latent concept space. Those can be fed into an addition model to extract semantic segments, bounding boxes etc. IIUC this is how dinov3 works.