For years Meta kept its best models to itself and gave the weaker ones away. On July 9 that ended. The company launched its first paid product — the Meta Model API — and put Muse Spark 1.1 behind it. Zuckerberg called the price “roughly 25% of what Anthropic and OpenAI charge.” Bloomberg, CNBC, and Reuters all ran with the price-war angle.
What it actually is
An agentic reasoning model sold as an API, not a chatbot. Muse Spark 1.1 handles a 1-million-token context window — a whole codebase in one session — and, unusually, it actively manages and compresses that context during long runs instead of just drowning in it. It does tool calls and computer use across desktop, mobile, and browser, and can spin off subagents to work in parallel. Coding agents are the obvious target.
The API entry point
Public preview, US developers only, $20 in free credits to start. Pricing is $1.25 input / $4.25 output per million tokens. The clever part: it’s drop-in compatible with the OpenAI and Anthropic SDKs, so swapping it into an existing agent stack is a few lines, not a rewrite. That’s the real threat — Meta made switching cheap on purpose.
You Might Also Like
- Cursor Composer 2 Takes on Anthropic and Openai With a 0 50 m Token Coding Model and the Benchmarks Back it up
- Fara 7b Microsofts Tiny Model That can Actually use Your Computer
- Openai Oauth Turns Your Chatgpt Subscription Into a Free Openai api but Should you use it
- Anthropic Gives Claude Computer use on mac and you can Control it From Your Phone
- Meta Muse Spark 14b and 9 Months Later Alexandr Wang Delivers but its Closed Source

Leave a comment