Meta releases Muse Wonder AI models built to run personal and professional agents
Posted
Duncan Riley (via SiliconANGLE)
Meta Platforms Inc. today released Muse Wonder and Muse Wonder Full, two artificial intelligence models it says are its most capable yet at running agents that handle personal and professional tasks. Meta aimed the release at the longer, multistep jobs people hand off to an AI agent. One example it gave was working through an […]
AI,
NEWS,
The-Latest,
Top Story 3,
Active Focus,
Active Memory,
AI agents,
AI models,
AI safety,
Alexandr Wang,
Anthropic,
Artificial Analysis,
benchmarks,
Claude Code,
Claude Fable 5.1,
Claude Opus 5.5,
Codex,
Coding,
cybersecurity,
DeepSWE,
ExploitGym,
GPT-6 Astra,
Harvey,
HealthBench Pro,
Meta,
Meta Model API,
Meta Superintelligence Labs,
Meta Superintelligence Scaling Framework,
Muse,
Muse Code,
Muse Spark,
Muse Spark 1.3,
Muse Wonder,
Muse Wonder Full,
OpenAI,
OSWorld,
ProgramBench,
Prometheus,
SWE-Marathon
