Loading video...

Video Failed to Load

Go Home

Meta is back with Muse Glimmer: a 30B open-source multimodal model built for local, agentic use. HF is shipping day-0 support and I built a few demos to see what it can do. First: we gave Glimmer tools and asked it to quantize itself.

ben_burtenshaw's profile picture

Ben Burtenshaw

9,000 subscribers

50,794 views • 12 days ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

1/ Today was the first time offline computer use felt real to us. We ran Muse Glimmer 30B from AI at Meta locally and watched it operate Notes and Reminders on macOS through Cua Driver We didn’t expect a 30B local model to get this far 🧵
0:26

1/ Today was the first time offline computer use felt real to us. We ran Muse Glimmer 30B from AI at Meta locally and watched it operate Notes and Reminders on macOS through Cua Driver We didn’t expect a 30B local model to get this far 🧵

Francesco

30,095 views • 11 days ago

Meta opened the weights this morning. It's already living on a Mac. Muse Glimmer 30B, running locally in Osaurus. It analyzed an image and built a playable game. Your hardware. Your model. No subscription.
0:40

Meta opened the weights this morning. It's already living on a Mac. Muse Glimmer 30B, running locally in Osaurus. It analyzed an image and built a playable game. Your hardware. Your model. No subscription.

Osaurus

23,103 views • 11 days ago

1/ muse spark 1.2 is a very strong multimodal model—it can do visual coding, robotics planning, and audio-visual understanding that all come together through agentic tools.
0:19

1/ muse spark 1.2 is a very strong multimodal model—it can do visual coding, robotics planning, and audio-visual understanding that all come together through agentic tools.

Alexandr Wang

142,275 views • 1 day ago

2-bit Muse Glimmer GGUF managed to call 100+ tools on just 14GB RAM. 🔥 Muse Glimmer did a complete repo bug hunt for 5 mins nonstop with: evidence, repro, fix, tests and a PR writeup. Run and train it in Unsloth. GitHub repo:
1:58

2-bit Muse Glimmer GGUF managed to call 100+ tools on just 14GB RAM. 🔥 Muse Glimmer did a complete repo bug hunt for 5 mins nonstop with: evidence, repro, fix, tests and a PR writeup. Run and train it in Unsloth. GitHub repo:

Unsloth AI

121,652 views • 12 days ago

I built Feynman, Claude Code for research. I gave it a question and it came back 30 minutes later with a cited meta analysis. It can also replicate experiments on Runpod, audit claims against code, and simulate peer review. Open source & MIT license, link below
0:26

I built Feynman, Claude Code for research. I gave it a question and it came back 30 minutes later with a cited meta analysis. It can also replicate experiments on Runpod, audit claims against code, and simulate peer review. Open source & MIT license, link below

Advait Paliwal

341,467 views • 4 months ago

I built a Nous Research Hermes workflow that my team at Box uses to track AI trends. Every day at 7am, it researches what's happening across the AI ecosystem and generates a brief for our team. We use it to identify: • AI trends worth paying attention to • Open source projects gaining momentum • New tools developers are adopting • Content opportunities to work on next I've been building agentic workflows for months, and this is one I've actually integrated into my team's daily workflow. Here's how I built it:
5:43

I built a Nous Research Hermes workflow that my team at Box uses to track AI trends. Every day at 7am, it researches what's happening across the AI ecosystem and generates a brief for our team. We use it to identify: • AI trends worth paying attention to • Open source projects gaining momentum • New tools developers are adopting • Content opportunities to work on next I've been building agentic workflows for months, and this is one I've actually integrated into my team's daily workflow. Here's how I built it:

Chris Kim

78,552 views • 1 month ago

Muse Glimmer can complete multi-step agentic tasks end-to-end from a single natural language prompt. In this demo, it autonomously discovers a local Home Assistant instance via network tool calls, queries device APIs, writes a responsive HTML/CSS/JS dashboard from scratch, and deploys a local server for verification.
1:54

Muse Glimmer can complete multi-step agentic tasks end-to-end from a single natural language prompt. In this demo, it autonomously discovers a local Home Assistant instance via network tool calls, queries device APIs, writes a responsive HTML/CSS/JS dashboard from scratch, and deploys a local server for verification.

AI at Meta

52,128 views • 12 days ago

This is my Figma to SwiftUI code workflow I couldn't find a good way to turn my Figma designs to Xcode so I built a plugin. It took roughly a day to create with Claude AI. A few more days to add additional features like Save file, Fast forward, Preview options, advanced settings, a ton of testing and of course shipping/feedback. We're at a point where we can built our own tools to complete a workflow.
14:33

This is my Figma to SwiftUI code workflow I couldn't find a good way to turn my Figma designs to Xcode so I built a plugin. It took roughly a day to create with Claude AI. A few more days to add additional features like Save file, Fast forward, Preview options, advanced settings, a ton of testing and of course shipping/feedback. We're at a point where we can built our own tools to complete a workflow.

Meng To

51,659 views • 1 year ago

First impressions on Muse Glimmer! It's incredibly fast for a dense model, currently running an average of 208tps with a max of 274tps on a single 5090 with their DFLASH config. Comparatively, though, both using Open Code, Qwopus Coder (with thinking off) produced a much better shark survival game than the one I got from Glimmer. Meta's new dense model is currently just lacking some HTML canvas taste, but this is something that can be added via SFT as long as the model is stable and capable from a back-end programming perspective. And it seems to be, without a doubt. The big kicker here is that I ran this at extra high thinking, and it did not take long at all to run. Our current local leader, Qwen 27B 3.6, has a tendency to overthink, but with glimmer, that is not the case. Right now, my recommendation for general local programming (Apps, Games, Websites, Visual Tools) in this class is still Qwopus Coder with thinking disabled, or Qwopus Fusion with thinking enabled. Of course Shark Survival is a very basic domain-specific test, but I find that the result scales very well across many domains. If we're going to be shipping apps generated entirely locally, visual taste is somewhat of a bare minimum requirement, solely in my opinion, and Qwen's models in this class offer significantly more at the moment. That's actually why I initially started getting into finetuning with Qwen 3.5, they were the first base that was able to do really good front-end with some opus-trace fine-tuning. Qwen 3.6 has taste even in the base model, and we know Qwen 3.8 is going to blow us all away! Regardless, this looks like a very tempting new base model. As a first offering from Meta in this class for a long time, I am incredibly impressed and elated to have it. We now finally have a proper Single GPU frontier race, instead of us just begging Qwen for more releases. Single GPU open frontier model race is a VERY good thing. Please keep pushing Meta
1:20

First impressions on Muse Glimmer! It's incredibly fast for a dense model, currently running an average of 208tps with a max of 274tps on a single 5090 with their DFLASH config. Comparatively, though, both using Open Code, Qwopus Coder (with thinking off) produced a much better shark survival game than the one I got from Glimmer. Meta's new dense model is currently just lacking some HTML canvas taste, but this is something that can be added via SFT as long as the model is stable and capable from a back-end programming perspective. And it seems to be, without a doubt. The big kicker here is that I ran this at extra high thinking, and it did not take long at all to run. Our current local leader, Qwen 27B 3.6, has a tendency to overthink, but with glimmer, that is not the case. Right now, my recommendation for general local programming (Apps, Games, Websites, Visual Tools) in this class is still Qwopus Coder with thinking disabled, or Qwopus Fusion with thinking enabled. Of course Shark Survival is a very basic domain-specific test, but I find that the result scales very well across many domains. If we're going to be shipping apps generated entirely locally, visual taste is somewhat of a bare minimum requirement, solely in my opinion, and Qwen's models in this class offer significantly more at the moment. That's actually why I initially started getting into finetuning with Qwen 3.5, they were the first base that was able to do really good front-end with some opus-trace fine-tuning. Qwen 3.6 has taste even in the base model, and we know Qwen 3.8 is going to blow us all away! Regardless, this looks like a very tempting new base model. As a first offering from Meta in this class for a long time, I am incredibly impressed and elated to have it. We now finally have a proper Single GPU frontier race, instead of us just begging Qwen for more releases. Single GPU open frontier model race is a VERY good thing. Please keep pushing Meta

Kyle Hessling

17,399 views • 11 days ago

A few weeks ago, I shared a Shopify store I built with v0 that you can explore like a game. People asked how I built it, so I recorded a demo to teach you how to build your own. I'd love to see what you create!
2:33

A few weeks ago, I shared a Shopify store I built with v0 that you can explore like a game. People asked how I built it, so I recorded a demo to teach you how to build your own. I'd love to see what you create!

Esteban Suárez

20,212 views • 1 month ago

“She’s the super villain to my superhero” 🫶 Is that a glimmer of friendship we see for Richard and Caroline? #BBUK
1:55

“She’s the super villain to my superhero” 🫶 Is that a glimmer of friendship we see for Richard and Caroline? #BBUK

Big Brother UK

14,894 views • 9 months ago

Thrilled to see Amazon Web Services making a major contribution to the open source AI community with the launch of the Strands Agents, an open source AI agents SDK! The core of Strands is the simple agentic loop that connects the model and tools together, like the two strands of DNA. This model-driven approach to agent building eliminates the need for complex agent orchestration by embracing the capabilities of state-of-the-art models to plan, chain thoughts, call tools, and reflect. Providing open source tools and interoperability with open source protocols is an important part of our strategy to enable an agentic future. Can't wait to see what you build with Strands!
1:33

Thrilled to see Amazon Web Services making a major contribution to the open source AI community with the launch of the Strands Agents, an open source AI agents SDK! The core of Strands is the simple agentic loop that connects the model and tools together, like the two strands of DNA. This model-driven approach to agent building eliminates the need for complex agent orchestration by embracing the capabilities of state-of-the-art models to plan, chain thoughts, call tools, and reflect. Providing open source tools and interoperability with open source protocols is an important part of our strategy to enable an agentic future. Can't wait to see what you build with Strands!

Swami Sivasubramanian

32,185 views • 1 year ago

When u use Fable / Opus to generate models it can be quiet a pain in the ass to do small iterations. I also needed a way for the model to easily visualise the results of what he built so he could iterate over it. So I built a custom model viewer it can access and where I can easily fine tune models manually if needed. Used it for my game Valoria
2:26

When u use Fable / Opus to generate models it can be quiet a pain in the ass to do small iterations. I also needed a way for the model to easily visualise the results of what he built so he could iterate over it. So I built a custom model viewer it can access and where I can easily fine tune models manually if needed. Used it for my game Valoria

moht

11,135 views • 2 months ago

Meta Tribe2 is a model that predicts your brain's response on videos so now you can edit videos with more clarity of how to compose them, what works and what doesn't and we built a free tool for you based on this model yes, totally free see details below
0:24

Meta Tribe2 is a model that predicts your brain's response on videos so now you can edit videos with more clarity of how to compose them, what works and what doesn't and we built a free tool for you based on this model yes, totally free see details below

AmirMušić

847,898 views • 3 months ago

Meet Waldo: Glean’s first agentic search model. Built on NVIDIA Nemotron 3 Nano and post-trained for search planning, Waldo figures out how to break down a query, which tools to call, what to read next, and when it has enough evidence to hand off.
0:50

Meet Waldo: Glean’s first agentic search model. Built on NVIDIA Nemotron 3 Nano and post-trained for search planning, Waldo figures out how to break down a query, which tools to call, what to read next, and when it has enough evidence to hand off.

Glean

98,637 views • 3 months ago

This is Repo Prompt 2.0 A fully integrated agent, that makes it seamless to use RP's powerful MCP tools, with a built-in oracle and context builder. A first class experience showcasing how much better and efficient your agents can be with good context engineering tools.
4:35

This is Repo Prompt 2.0 A fully integrated agent, that makes it seamless to use RP's powerful MCP tools, with a built-in oracle and context builder. A first class experience showcasing how much better and efficient your agents can be with good context engineering tools.

eric provencher

38,921 views • 5 months ago

We built the first open‑source AI glasses with a brain interface They use OmiGlass with an fNIRS chip by Telepathic Technologies They can record your entire day and track your focus or intent This is just the beginning => we will make it awesome All links below
0:56

We built the first open‑source AI glasses with a brain interface They use OmiGlass with an fNIRS chip by Telepathic Technologies They can record your entire day and track your focus or intent This is just the beginning => we will make it awesome All links below

Nik Shevchenko

25,677 views • 1 year ago

This beautiful presentation is by Muse 1.2. It used gpt-image-v2 to create the images by itself. I told it to make a history page from Llama to Muse 1.2, and it did a great job. Alexandr Wang , you guys created magic with agentic tasks. I'm now waiting for the big model from you guys.
1:10

This beautiful presentation is by Muse 1.2. It used gpt-image-v2 to create the images by itself. I told it to make a history page from Llama to Muse 1.2, and it did a great job. Alexandr Wang , you guys created magic with agentic tasks. I'm now waiting for the big model from you guys.

Chetaslua

17,158 views • 15 days ago

Today we're shipping Nemotron 3 Ultra. A 550B MoE frontier-intelligence open model built for long-running agents. It delivers 5x faster inference and lowers the cost of complex agentic tasks by up to 30% versus other open frontier models.
2:59

Today we're shipping Nemotron 3 Ultra. A 550B MoE frontier-intelligence open model built for long-running agents. It delivers 5x faster inference and lowers the cost of complex agentic tasks by up to 30% versus other open frontier models.

NVIDIA AI

1,327,029 views • 2 months ago