Fieldbook to AIThe plain-English guide to which AI to use

All models

Muse Spark Open

A downloadable Meta model tuned for programming. The one to watch if you want to run your own coding helper.

Last updated .

Costs here use the prices AI companies charge software developers, so they show relative cost, not your monthly bill. See what ChatGPT, Claude and Gemini charge you.

Made by Meta, the company behind Meta AI. Free to download.

What it costs

Based on list prices and the assumptions on How we work. Your real cost will differ.

At a glance

App plans and prices for Meta AI: see plans.

Best for

running your own coding helper

Watch out for

Brand new (listed Oct 15, 2026) with no smartness score yet. Check the computer requirements before planning to run it yourself.

Appears in

Frequently asked

How much does Muse Spark Open cost to use?
A 1,000-word essay costs about $0.01 (about 1 cent); summarizing a 50-page PDF about $0.04 (about 4 cents). These are developer (API) prices, not an app subscription.
What is Muse Spark Open best for?
A downloadable Meta model tuned for programming. The one to watch if you want to run your own coding helper.
How much can Muse Spark Open read at once?
About 1,570 pages (a rule of thumb: 1,048,576 tokens at roughly 500 words a page).
Can I download Muse Spark Open for free?
Yes, the model files are published, though running a large one needs powerful hardware.

Under the hood

MakerMeta
Price per 1M input tokens$1.25
Price per 1M output tokens$4.25
Context size1.05M tokens (1,048,576)
Artificial Analysis Intelligence IndexNot reported (source, via OpenRouter)
Open weightsYes (Hugging Face: meta-models/Muse-Spark-Open-816B-A42B)
SpeedNot measured yet
Listed2026-10-15

Maker's description (technical)

Muse Spark Open is a coding-optimized open-weight model. Trained specifically for agentic coding workflows, it delivers higher first-attempt accuracy, reliable tool calling, and deep codebase understanding across the repos developers...

View on OpenRouter. Our verdict above is editorial, not benchmarked by us.