Hiero

Hiero works in Language, and is recorded as handling translation. These are the areas it was designed around; they describe intent rather than a hard boundary.

None. Hiero is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

The licensing for Hiero was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

Hiero has 120M parameters. Very unsure, but the paper mentions "We ran the training process of Section 3 on the same data, obtaining a grammar of 24M rules" and "For our experiments we used the following features, analogous to Pharaoh’s default feature set: • P(γ | α) and P(α | γ), the latter of which is not found in the noisy-channel model, but has been previously found to be a helpful feature (Och and Ney, 2002); • the lexical weights Pw(γ | α) and Pw(α | γ) (Koehn et al., 2003), which estimate how well the words in α translate the words in γ; 2 • a phrase penalty exp(1), which allows the model to learn a preference for longer or shorter derivations, analogous to Koehn’sphrase penalty (Koehn, 2003)." Suggesting 24M rules * 5 features per rule (?). That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

Hiero was published by University of Maryland, based in United States of America, categorised as academia.

Hiero was published in June 2005. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.