Plan event · 12 Nov 2026

Muse Codemodels

What moved

Muse Ember was stored on no muse-code plan — the four rosters held Muse Wonder (plus Muse Wonder Full on Power Usage and Maximum Usage) and the Muse Spark 1.3 / 1.2 variants only → Muse Ember is stored on all four plans — Everyday Usage, High Usage, Power Usage and Maximum Usage — priced at the Standard-mode ≤ 200K schedule ($0.004 cached input / $0.10 input / $0.50 output per 1M), with the > 200K long-context band ($0.012 / $0.30 / $1.50) in the rule’s notes; no Contributor variant exists

Muse Ember was missing from every stored roster even though both docs pages already grant it: dev.meta.ai/docs/muse-code/subscriptions names it on the Everyday Usage feature card — "Access to Muse Ember, Muse Spark, Muse Wonder and Muse Image" — with High, Power and Maximum inheriting it through the "Everything from the ... plan" cascade, and dev.meta.ai/docs/muse-ember carries the per-model availability table row "Muse Ember (muse-ember-1.0) | Standard mode | Pay-as-you-go and every plan" with the prose "Muse Ember is on Model API and in Muse Code, on pay-as-you-go and with any subscription." The marketing card on dev.meta.ai/products/muse-code still omits Ember from its Everyday feature list ("Access to Muse Spark, Muse Wonder and Muse Image") — recorded here as a conflict and resolved in favour of the per-model availability table, the stronger source. The calculator prices the new row from the published two-tier schedule: up to 200K input tokens per request $0.004 cached input / $0.10 input / $0.50 output per 1M (the cheap band stored), over 200K $0.012 / $0.30 / $1.50 billed in full for the request — and the muse-ember page adds "On a Muse Code subscription, a request with more than 200K input tokens uses your plan's usage faster." Ember runs in Standard mode only on Model API ("it has no Contributor-mode variant"), so no Contributor row is created. The model card publishes input "Text, image, video, audio, and PDF", output "Text", a 1,048,576-token context window and 524,288 max output. Everything else re-read unchanged: plan prices $5 / $15 / $50 / $200 per month, the 10–50 prompts/5h limits and the 5× / 20× / 80× ceilings (50 / 250 / 1000 / 4000 requests per 5h), Muse Wonder on every plan, Muse Wonder Full on Power Usage and Maximum Usage only; Muse Image stays out of scope (image generation) and Muse Spark 1.1 stays unavailable in Muse Code.

Verified 2026-11-12 · source

Muse Codemodels

Affected plans

More Muse Code events

Never miss a plan churn — get the weekly brief.