Meta Releases Muse Ember Lightweight AI Model Optimized for Agentic Tasks
Meta's Superintelligence Lab has launched Muse Ember, its most cost-efficient and lightweight AI model to date. Designed specifically for high-volume agentic workflows—such as tool calling, coding subtasks, and document parsing—Ember is positioned to work in tandem with Meta's larger 'Muse Wonder' model to balance complex reasoning with scalable execution. The model offers significant cost advantages, priced at 10 cents per million input tokens. This makes it more competitive than frontier models like Anthropic’s Haiku and OpenAI’s Luna, as well as open-source alternatives like DeepSeek V4.1 Flash. Additionally, Ember features native multi-modal capabilities, allowing agents to process text, images, audio, and video without requiring separate specialized models.
Tickers: $META
Topics: Artificial Intelligence Large Language Models Meta Superintelligence Lab
Read the full article at TechCrunch
Part of the story Meta Platforms Introduces Low-Cost Muse Ember AI Model for Developers (4 articles)