---
id: "2026-11-09-meta-muse-ember"
url: "https://postcutoff.com/e/2026-11-09-meta-muse-ember/"
as_of: "2026-11-13T13:49:00+01:00"
date: "2026-11-09"
date_precision: day
category: model-release
importance: 3
confidence: high
status: [Confirmed]
sources: 3
editor: human editor
human_review: null
version: "2026-11-10"
---

As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/e/2026-11-09-meta-muse-ember/

# Meta releases Muse Ember, its smallest and cheapest Muse model, for agent workloads

Meta Superintelligence Labs released Muse Ember 1.0 on 9 November 2026, a small reasoning model for high-volume agent work (tool calls, routing, extraction) priced at $0.10/$0.50 per million tokens for prompts up to 200K tokens, with a 1M-token context. Artificial Analysis puts it around 34-36 on its Intelligence Index, on the cost-efficiency frontier.

## Key facts
- Price: $0.10 input / $0.50 output per 1M tokens up to 200K-token prompts ($0.004 cached input); $0.30 / $1.50 above 200K
- Up to 1M input tokens; text, image and audio input; six reasoning levels (minimal to max)
- Meta: 'Muse Wonder plans, Muse Ember handles the volume'; claims it beats Claude Haiku and GPT-6 Luna on the AA index per cost
- Artificial Analysis Intelligence Index about 34 (xhigh) and 36 (max), roughly $0.12-0.14 per index task (read from AA's chart by Simon Willison)
- Simon Willison: same short-prompt price as GPT-6 Luna and Claude Haiku 5.5; its pelican SVGs comparable to Muse Wonder's at about 1/40 the price

## What happened

Meta added a low-cost tier below Muse Spark and Muse Wonder, aimed at the many cheap calls an agent system makes. The model id is
`muse-ember-1.0`. TechCrunch did not report open weights.

## Why it matters

It puts Meta in the cheap "workhorse" tier against GPT-6 Luna, Claude Haiku 5.5 and Chinese flash models, where most agent tokens are spent.

## Your AI and this story

- GPT-6 Astra (training cutoff April 2026): 193 days after its cutoff
- Claude Opus 5.5 (training cutoff June 2026): 132 days after its cutoff
- Gemini 4 Hydrogen (training cutoff March 2026): 223 days after its cutoff
- Grok 4.7 (training cutoff May 2026): 162 days after its cutoff

## Sources

1. [Meta: Introducing Muse Ember](https://research.meta.ai/blog/muse-ember) (research.meta.ai, official)
2. [TechCrunch: Meta releases Muse Ember](https://techcrunch.com/2026/11/09/meta-releases-muse-ember-a-new-lightweight-model-for-agent-execution/) (techcrunch.com, press)
3. [Simon Willison on Muse Ember](https://simonwillison.net/2026/Nov/10/muse-ember/) (simonwillison.net, discussion)

## Changes

- 2026-11-10 (filed): Created

## Related

- 2026-10-12: [Meta releases Muse Wonder and Muse Wonder Full, its first frontier-class models](https://postcutoff.com/e/2026-10-12-meta-muse-wonder/index.md)
- People: [Simon Willison](https://postcutoff.com/person/simon-willison/)
- Models: [Muse Ember 1.0](https://postcutoff.com/m/muse-ember-1-0/) (`muse-ember-1-0`)