# Sam Witteveen — 'Meta's Open Weight: Muse Glimmer 30B'

> Sam Witteveen walks through Meta's Muse Glimmer, a 30B Apache-2.0 agentic model built to run on one consumer GPU.

- Org: Sam Witteveen
- Date: 2026-08-10
- Category: video
- Source: https://www.youtube.com/watch?v=Wjh6wx2dl3Q
- Canonical page: https://ai-tldr.dev/releases/sam-witteveen-muse-glimmer-aug10/

## In short

Sam Witteveen's new video walks through Muse Glimmer, the ~29.6B Apache-2.0 agentic model Meta released today. It takes text and images, holds 131,072+ tokens of context, and fits in 17-20 GB once quantized.

## What it is

Muse Glimmer is the ~29.6B-parameter open-weight model Meta published on 2026-08-10, and this video is Sam Witteveen's same-day walkthrough of it. The model is Apache-2.0, takes text and images as input, and is aimed at always-on local agent work rather than cloud serving.

## How it works

The architecture is a dense causal transformer with a separate ~1.8B perception encoder that turns each image into up to 4,096 visual tokens. Context runs to 131,072+ tokens, and a bundled DFlash drafter does speculative decoding for a 3.1x speedup on an RTX 5090 and 1.5x-1.8x on Apple silicon.

## Why it matters

Quantized builds bring Muse Glimmer down to 17-20 GB, so an RTX 5090 or an M4/M5 Max Mac can run the whole agent loop with no API bill. Sam Witteveen's channel is where many local-LLM builders decide what to download first, so a same-day review shapes how fast the weights spread.

## Who it's for

local-LLM builders, agent developers

## Try it

Watch the video, then pull `meta-models/Muse-Glimmer-30B` from Hugging Face

## Links

- [Watch on YouTube](https://www.youtube.com/watch?v=Wjh6wx2dl3Q)
- [Sam Witteveen Channel](https://www.youtube.com/channel/UC55ODQSvARtgSyc8ThfiepQ)
- [Meta Announcement](https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model)
- [Model Card](https://huggingface.co/meta-models/Muse-Glimmer-30B)