Meta Muse Ember: smallest, cheapest Muse model for AI agents at $0.10 per million input tokens

Meta Superintelligence Labs released Muse Ember on November 9, 2026, its smallest and most efficient model, designed for complex reasoning execution and the work agents repeat all day: tool calls, coding subtasks, routing, extraction, and long document parsing. It is available today on Muse Code and the Meta Model API as muse-ember-1.0, supports the Responses API, and natively understands text, images, audio, and video. Ember is the first model Meta trained using Muse Wonder Full, and Meta pitches the pairing directly: Muse Wonder plans, Muse Ember handles the volume.

On Artificial Analysis benchmarks dated November 9, 2026, Muse Ember at max reasoning scores 36 on the Intelligence Index, matching Claude Haiku 5.5 at max and ahead of GPT-6 Luna at max with 28. Ember leads both rivals on AutomationBench-AA end-to-end business workflows (37.8 against 13.9 and 20.2), Agents' Last Exam (61.6 against 57.0 and 57.1), and Terminal-Bench 4.0 agentic coding (40.9 against 39.2 and 16.4). Meta puts Ember's cost at about $0.14 per Artificial Analysis task, against $2.39 for Muse Spark 1.3, which scores 37 on the same index.

Ember costs $0.10 per million input tokens, $0.004 per million cached input tokens, and $0.50 per million output tokens, with a 1M-token context window and up to 512K output tokens, so it can take in a full codebase or document set in one call. Its reasoning_effort setting runs from minimal to max, and Meta evaluated it before release under its Superintelligence Scaling Framework, finding it below the high-risk thresholds in every category assessed. Early partner Box uses it to score contracts against negotiation playbooks and to screen candidates against detailed briefs.