Meta launches the lightweight AI model Muse Ember: benchmark scores rival Claude Haiku 5.5, with per-task costs just one-third of the price.
Muse Ember is designed for complex reasoning and agents, supporting a 1 million-token context window to process entire codebases in one go. It natively understands text, images, audio, and video. Its benchmark scores exceed GPT-6 Luna, with pricing at just $0.10 per million input tokens. #Meta AI model#
Continue looking at this topic
OpportunitiesMastering AI + Short Video Matrix for Customer Acquisition in 30 Days: From Scratch to Solo Operation ↗News and AnalysisCan Claude Code handle Conductor workflows? ↗Open-source projectHow to choose the right Claude Code terminal agent? ↗Hotly discussed viewsAI is tasked with creating educational videos based on implementation plans: how do we avoid scenarios where the visuals look great but the information is wrong? ↗Sources of the same event
- Meta launches the lightweight AI model Muse Ember: benchmark scores rival Claude Haiku 5.5, with per-task costs just one-third of the price.
- Meta launches the lightweight AI model Muse Ember. Its benchmark scores rival Claude Haiku 5.5, while the cost per task is just one-third.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open the original article on ITHome