NodeValid

Published video-model weights, licences and hardware notes

Which vendors publish an exact parameter count

One of the 15 families publishes a parameter count to the unit. One publishes a bound with no upper end, two publish component breakdowns, four publish counts across several variants, six publish a single round figure and one publishes nothing. As of 2026-09-12.

Five shapes the size column takesA count can be measured, bounded, decomposed, spread across variants or simply described, and only some of those shapes support a comparison between two rows.Supports a direct comparisonExact countA measurement of one file, with no convention question leftopen.Round figureA description; comparable within an order of magnitude.Component splitTwo numbers that name the architecture as well as the size.A boundA floor with no ceiling, so nothing can be placed beside it.Comparable, or merely indicative
Fig. 1 The shapes at the top can be compared with each other; the ones below cannot.
The shape of the size column across the register, by how the count is published. Recorded 2026-09-12.
ItemWhat is publishedDetail
An exact countOne family2,059,174,912 parameters, published on the model card
A boundOne familyOver 13 billion parameters, with no upper end stated
Components published apartTwo familiesVAE 175M and DiT 2.8B; 10 billion plus a 362M video autoencoder
Counts across variantsFour families2B and 5B; 1.3B, 5B and 14B; 24B and 4.5B; 13B with distilled and fp8 builds
A single round figureSix families30 billion, 11B, 13B, 2B, 13.6B and a 5B
No count at allOne familyThe card states clip lengths and resolutions instead

Inclusion rule. One row per shape the count takes. A shape used once keeps its row, because a single exact count is what makes the rest of the column legible as approximations. Order. From the most precisely stated shape to the absence of one.

1Exactness removes the convention question

A rounded two billion could include or exclude a text encoder and a reader would never know. A count to the unit is a statement about one artefact, which is why one row here serves as the reference the rest of the column is read against.

It does not make the row more useful in every way. The same card still describes one configuration, and precision does not extend a figure to a different clip length or frame size.

2A bound is defensible and uncomparable

Counting conventions differ between publishers, and writing a bound avoids committing to one. That is reasonable for an author and it removes the row from the one comparison the size column exists to support.

The register keeps the phrasing. Rounding a bound into a number would manufacture a figure, and treating it as an exact count would manufacture a comparison.

3Component splits carry the most information per character

Two labelled numbers name the architecture while stating the size, so a reader who recognises the labels knows roughly how the run will behave before touching it. A total would be arithmetically true and less useful.

It also shows which component a memory figure is dominated by, which is the practical question a count is usually asked in place of.

4Sources

Repository statements and hosted rates are linked from the model pages and the how read page, checked 2026-09-12. Hardware figures are set out on requirements. Nearby: Complete output, Two figures, one run, Which variant.