Meta unveiled Muse Ember on Monday, November 9, 2026, its smallest and most efficient AI model to date. It is specifically designed for the repetitive tasks that AI agents perform throughout the day, such as tool calls, coding subtasks, routing, data extraction, and long document parsing.
What Muse Ember offers
According to Meta, Ember delivers more intelligence per dollar than Anthropic’s Claude Haiku 5.5 or OpenAI’s GPT-6 Luna at any reasoning setting. It matches Haiku’s best score at about a third of the cost and beats every Luna configuration. The model sits on the efficiency frontier of the Artificial Analysis leaderboard.
Pricing is $0.10 per million input tokens and $0.50 per million output tokens. It features a 1 million token context window and can generate up to 512,000 output tokens. It is natively multimodal, processing text, images, audio, and video without separate models.
Benchmark performance
On the Artificial Analysis Intelligence Index, Ember scores 36 at maximum reasoning, matching Haiku 5.5 but at roughly $0.14 per task versus $0.41 for the rival. In agent and coding tests, it leads on several metrics, though Haiku performs better on some real engineering tasks like FrontierCode.
Meta suggests pairing Ember with larger models such as Muse Wonder: Wonder handles complex planning while Ember executes the high-volume repetitive work.
Availability and context
The model is available today on the Meta Model API and Muse Code as muse-ember-1.0. It is the fourth model in the Muse lineup after Spark, Glimmer, and Wonder, and the first trained using Muse Wonder Full.
Partners including Box, Cloudflare, and Oracle have tested it on enterprise tasks and report strong accuracy, speed, and token efficiency.
By Grok
