Meta AI Releases Muse Code (Beta): A Terminal Coding Agent Powered by the New Muse Spark 1.2 Model
Meta AI has released Muse Code (in beta), a terminal coding agent in beta, powered by its new Muse Spark 1.2 model. Meta positions the pair as its next step toward the frontier, with larger models on the way. Muse Code targets complex software engineering across large repositories: it plans changes, writes code, and validates the results. A set of async background agents stays alive for the whole session instead of spawning per task. A local append-only event log records every model call, tool
Get smart on it
Meta AI has released Muse Code, a terminal coding agent that automates complex software engineering tasks like planning changes, writing code, and validating results across large code repositories. The agent uses persistent background workers that remain active throughout a session rather than spawning new ones for each task, and records all operations in a local event log that allows it to resume precisely after crashes. Muse Code is powered by Muse Spark 1.2, a new model trained specifically for coding work on long-horizon, repository-scale projects through co-training with the tool itself. The tool is available in beta for macOS and Linux and targets teams in software, developer tooling, and infrastructure engineering who need to perform large-scale code refactors, migrations, and optimizations.
Related news
Pokee AI Releases Pokee-Isaac 28B: A 10M-Token Context Agentic Model Built to Run Inside the Customer Boundary
Long-horizon agents accumulate context faster than they resolve tasks. Every tool output, observation, and intermediate reasoning step stays in the window, and the two capabilities that matter — holding that context and staying coherent across it — have so far been available almost exclusively from cloud endpoints. That excludes regulated industries, public-sector institutions, and on-device applications, where the data is not permitted to leave the boundary at all. Pokee AI released Pokee-Isaa
Mistral AI Releases Shieldstral 1.0 3B: An Open-Weights Policy-Adaptive Multimodal Safety Classifier Matching Models 7× Its Size
Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Most guardrail models bake their category list into the weights, so re-targeting one to a new deployment context means retraining — and the same content can be acceptable on a cybersecurity research tool while being harmful on a mental-health platform. Shieldstral inverts that: operat
