← Tuesday, October 20, 2026Models & trainingNews

AINews roundup reports an 816B-parameter open-weight model

latent.space 25 min read Editor score 6/10

The roundup describes Muse Spark Open as an 816B-total, 42B-active mixture-of-experts model with multimodal inputs and a one-million-token context. Those details help engineers assess a notable open-weight release, though the item is a newsletter roundup.

What

Muse Spark Open is an open-weight, 816B-total, 42B-active mixture-of-experts model with a 1M context.

How

Its NVFP4 checkpoint quantizes only the experts, reducing weights from about 1.6TB to about 530GB.

Why it matters

Meta says the quantized model fits on four Blackwell Ultra GPUs, eight H200s, or two 512GB M5 Ultra Mac Studios.

Read on latent.space →

Headline, summary, brief and diagram are written by a model from the source's own text; the link, byline and numbers come from the source. Read the original before relying on a claim.