Muse Spark 1.2 Pricing

Muse Spark 1.2 pricing is set at two tiers, released with Meta's 2026-08-05 launch. Standard costs $1.25 in and $4.25 out per million tokens. Contributor costs $0.10 in and $0.20 out per million tokens.

This page shows the full rate card, explains the Contributor tradeoff, and walks through a session-cost example so you can predict spend before you commit.

Every number here is quoted directly from the Meta launch email. Nothing has been rounded, estimated, or inferred.

The Full Rate Card

Muse Spark 1.2 has two tiers on identical model access — the only difference is whether Meta uses your data to improve their products.

Standard tier is $1.25 per million input tokens, $0.15 per million cached input tokens, and $4.25 per million output tokens. Meta does not use Standard prompts or outputs.

Contributor tier is $0.10 per million input tokens, $0.002 per million cached input tokens, and $0.20 per million output tokens. Meta uses Contributor prompts and outputs to improve their products.

• Standard input: $1.25 / M tokens
• Standard cached input: $0.15 / M tokens
• Standard output: $4.25 / M tokens
• Contributor input: $0.10 / M tokens
• Contributor cached input: $0.002 / M tokens
• Contributor output: $0.20 / M tokens

The Contributor Tradeoff

Contributor cuts your bill by roughly 12x on input and roughly 21x on output, in exchange for Meta using your prompts and outputs to improve their products.

That data-use clause is the whole difference. If your prompts include client code, PII, or trade secrets, Standard is the only defensible choice.

For personal projects, OSS work, and non-sensitive prototypes, Contributor is a straight win. When we run our own amazon-gear-radar routine on public product data, a tier like this is exactly where non-sensitive workloads belong.

• 12x cheaper input, 21x cheaper output
• Meta uses Contributor prompts + outputs (Standard: no)
• OSS and prototype workloads fit Contributor
• Client code and PII stay on Standard

Session-Cost Example: A PR Review

A single Muse Code PR-review session that reads a 40k-token diff and writes 6k tokens of review costs $0.076 on Standard ($1.25 * 0.040 + $4.25 * 0.006) and $0.005 on Contributor.

Scaled to 200 PR reviews per month, that is $15.20 on Standard versus $1.00 on Contributor. The gap widens sharply on output-heavy tasks like scaffolding or refactoring.

Cached input matters when the same context repeats across a session. At $0.15 (Standard) or $0.002 (Contributor) per million tokens, cache hits are effectively free — design your prompts so the shared context sits at the top.

• Per-PR review: ~$0.076 Standard vs ~$0.005 Contributor
• 200 PRs/mo: ~$15 vs ~$1
• Output-heavy tasks widen the gap
• Cache-friendly prompt structure saves real money

What Meta Has Not Published

Meta has not published rate limits (RPM/TPM), context window size, output-token caps, or free-trial credits at launch — verify at https://developer.meta.com/ai/products/muse-code/.

Meta has also not clarified whether Contributor data-use is training-only or covers evaluation, red-teaming, and product analytics too.

Enterprise pricing, volume discounts, and SLA-backed contracts are not documented in the launch materials. Contact Meta directly for enterprise terms.

• Rate limits: not published
• Context window: not published
• Free trial credits: not published
• Enterprise / SLA pricing: not documented

How to use Muse Spark 1.2

You do not host Muse Spark 1.2 yourself — you use it through a tool, so "getting started" really means choosing the right one.

The fastest way to put Muse Spark 1.2 to work day to day is inside an AI IDE, and Cursor is the most popular — it supports it directly, so you can be working in minutes. Prefer a different editor? Windsurf, Zed, and GitHub Copilot drive these models too.