Meta Launches Muse Glimmer – New Open AI Model for Autonomous Agents
Meta Superintelligence Labs has unveiled Muse Glimmer, an open AI model with 30 billion parameters, tailored for autonomous agent tasks on a single consumer graphics card.
What happened?
Meta Superintelligence Labs has launched Muse Glimmer, an open AI model featuring 30 billion parameters, developed specifically for autonomous agents. The model is released under the Apache 2.0 open-source licence. Through optimisation techniques, the memory footprint has been reduced from approximately 60 GB to between 18 and 20 GB, allowing it to run on a single consumer graphics card with 24 GB or 32 GB of VRAM, leaving space for runtime memory buffers.
Key facts
| Lanseringsdatum | 10 augusti 2026 |
|---|---|
| Antal parametrar | 30 miljarder |
| Minneskrav | 18–20 GB (passar 24/32 GB GPU) |
| Licens | Apache 2.0 |
Why it matters
Most open-source language models are trained for general text generation and dialogue, often causing them to lose focus during lengthy multi-step processes. Muse Glimmer employs an agent-oriented architecture designed specifically for 'agentic task completion'—the ability to autonomously plan, invoke external tools, troubleshoot when tools fail, and execute complex tasks over extended periods.
Who is affected?
The launch is primarily aimed at AI developers, researchers, and small businesses seeking to build autonomous agent applications locally without relying on proprietary cloud-based services. Hobbyists with high-end consumer graphics cards are also expected to benefit.
Impact on the EU
Model weights are available for download in the EU under the Apache 2.0 licence. This licensing model allows European developers to run the model locally on their own servers, eliminating the need to transfer data to the US.
What else you should know
It is worth noting that the 'first of its kind' claim primarily pertains to the combination of an agent-specific focus, open licensing, and optimisation for individual consumer GPUs. General-purpose models like Llama and Mistral have previously been used for agent tasks, but often required greater computational resources or extensive fine-tuning.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilken licens gäller för Muse Glimmer?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Decide whether this affects strategy over 6–12 months or is just noise.
- Discuss with leadership: do we own the right question or does ownership need to move?
- Ask: what risk are we taking by NOT acting on this this quarter?
Generated angle — not editorial analysis of "Meta Launches Muse Glimmer – New Open AI Model for Autonomou"