The 3D workflows caught my attention here. Coding and computer use become much more powerful together: write code, inspect the result inside the application, and keep refining it. Excited to see Kimi pushing toward complete creative workflows.View quoted post on X
A few months ago we set ourselves the goal of making an open model for the claw: good for agentic, fits in a 32GB gpu. Here's what went into Muse Glimmer: 1/?
x.com/AIatMeta/statu…View quoted post on X
Today we published a technical report on Muse Glimmer, our 30B open-weight model built for agentic tasks on consumer hardware.
It runs at practical speeds on a Mac or PC with a single consumer GPU at 4-bit precision, and leads comparably-sized open models on tool use and multi-step planning.
Read about the model architecture, distillation recipe, quantization and more: go.meta.me/4BcvxolView quoted post on X
1/ muse spark open is the most capable US open-weight model to date.
816B total params, 42B active. 1M-token context. natively multimodal. built for coding and agentic work like long-horizon tasks, hard software engineering, multi-step reasoning.
america! 🇺🇲View quoted post on X
From 🥑 to 🍉 in 6 months, what a journey! Yet more challenging tasks to work on!View quoted post on X
