Observed Signal · Nov 9, 2026 · Technical Release · Source: techcrunch · Impact: 4/5 · Sentiment: Positive

Meta launches Muse Ember, cheap model for agent execution

Executive Signal Summary

Meta Superintelligence Lab released Muse Ember on Monday, its lightest and most cost-efficient AI model yet. Priced at 10 cents per million input tokens, Ember is cheaper than frontier competitors like Anthropic's Haiku and OpenAI's Luna, while also being cost-competitive with open-source models like DeepSeek V4.1 Flash and GLM 5.3 Flash. Designed for agentic workloads, Ember handles tool calls, coding subtasks, routing, extraction, and long document parsing. It is intended to pair with larger models like Muse Wonder in an agentic harness: Muse Wonder plans, while Muse Ember handles the volume. The model also processes text, images, and audio, enabling agents to read screenshots, charts, and scanned documents, or work from voice memos and meeting recordings. Ember is the fourth model in Meta's Muse family, following Spark, Glimmer, and Wonder.

Polaris7 AgentPolaris7 Strategic Assessment
High Confidence

Meta's release of a highly cost-efficient and multimodal AI model specifically designed for agentic workloads signals a strategic push into the agentic AI space, which is expected to transform digital advertising and marketing automation. The low price point could democratize access to AI agents, impacting how agencies and platforms operate.

MARKET SIGNALS

Track Meta Signals & Strategic Shifts

Polaris7 connects regulatory filings, primary sources, leadership changes, and M&A activity into structured intelligence. Set up your Polaris7 workspace to monitor relevant companies in your executive briefing.

  • Verified primary sources
  • Automated briefings
  • Instant watchlist setup

Key Takeaways & Evidence Grounding

  • Meta released Muse Ember, its lightest and most cost-efficient AI model, on Monday.
  • Muse Ember costs 10 cents per million input tokens, cheaper than frontier competitors.
  • Ember is designed for agentic workloads like tool calls, coding subtasks, routing, and extraction.
  • It is multimodal, handling text, images, and audio.
  • Ember is the fourth model in Meta's Muse family, after Spark, Glimmer, and Wonder.

Connected Companies & Entities

“Meta Superintelligence Lab released a new AI model on Monday called Muse Ember, the lab's lightest and most cost-efficient AI model yet.”

“Meta particularly emphasized Ember's cost-per-task, using Artificial Analysis's composite benchmark score to show Ember as cheaper and smart...”

“Meta particularly emphasized Ember's cost-per-task, using Artificial Analysis's composite benchmark score to show Ember as cheaper and smart...”

“At 10 cents per million input tokens, Ember is significantly cheaper than frontier competitors like Haiku and Luna. Not only that, it’s cost...”

Ontology Mapping & Concepts

Primary Source Grounding & Direct Attribution
Direct Origin Attribution
Primary Reporting: techcrunch•Published: Nov 9, 2026
Original Coverage Title: “Meta releases Muse Ember, a new lightweight model for agent execution”

Related Market Signals & Shifts

Recent verified developments and strategic activity across this market segment.

AI / LLMNov 10, 2026

Meta Launches Muse Ember Model, Brings Muse to Apple Watch

Meta has unveiled Muse Ember, its smallest and most cost-efficient AI model designed for high-token, repetitive agentic tasks like coding sub-tasks, tool calls, and document analysis. The model is available via the Meta Model API and in Muse Code subscriptions, and it outperforms comparable models like Claude Haiku 5.5 and GPT-6 Luna on benchmarks such as Terminal-Bench 4.0 and Agents' Last Exam, while offering competitive pricing (starting at $0.10 per million input tokens). Meta also announced that its Muse agent is now available on Apple Watch and Android Wearable OS, following its initial launch in September 2026 and rapid adoption. The agent can be used hands-free via voice mode and competes with other AI assistants like dots, Claude Duet, Grok Bot, and Gemini.

Read assessment
Large Language Models (LLM) & AIJul 10, 2026

Meta unveils Muse Spark 1.1 agent model

Meta released Muse Spark 1.1, an agent-focused multimodal model designed to plan tasks, coordinate parallel AI sub‑agents, use external tools and execute end‑to‑end workflows. Muse Spark 1.1 supports images, video and audio, claims a context window up to one million tokens, and emphasizes visual‑to‑code generation and agentic control of desktop/web interfaces. Meta published internal benchmarks placing Muse Spark 1.1 competitive with current frontier models (GPT‑5.5, Claude Opus 4.8, Google Gemini 3.1 Pro) for agents and coding tasks. Meta also published API pricing (USD 1.25 per 1M input tokens; USD 4.25 per 1M output tokens) and offers $20 trial credit for new accounts. Demos shown include automating a Facebook Marketplace listing and debugging/code migration workflows, and Meta says developers and researchers already use the model for daily development and research tasks.

Read assessment
AIOct 12, 2026

Meta Unveils Frontier Models Muse Wonder

Meta Superintelligence Labs (MSL) launched its second-generation frontier AI models, Muse Wonder and Wonder Full, featuring one-million-token context windows and Active Memory/Active Focus capabilities. Independent benchmarks (Artificial Analysis Index v4.3) rank Wonder Full second (score 56) behind Claude Opus 5.5 (58); Meta claims parity with GPT-6.1 Sol and Claude Fable 5. Pricing is $4/$20 per million input/output tokens, with Wonder available via API and Muse Code, while Wonder Full is Muse Code-exclusive. Additionally, Meta released Muse Spark Open (816B parameters, 42B active), positioned as the best non-Chinese open-weight model, optimized for Apple's Mac Studio. Meta's strategic pivot includes a $160 billion investment and monetization via transaction fees. A launch-day usage-limit bug was fixed with a quota reset. In other AI news: CoreWeave faced debt concerns (shares -15.5%), Character.AI filed for Chapter 11, Modal raised $750M Series D, and Anthropic's Q3 revenue hit $16.8B.

Read assessment

WATCHLIST & BRIEFINGS

Receive future signals for Meta & competitors in your briefing

Stay ahead of material M&A deals, leadership changes, account wins, and product rollouts. Polaris7 connects primary filings, company newsrooms, and trade reporting directly into your executive briefing.

  • Regular primary-source checks
  • Structured Watchlist briefings
  • Preserves current context