Meta Rises 4% as Muse Spark 1.3 Claims Parity With Anthropic and OpenAI

Meta just fired its loudest shot yet at Anthropic and OpenAI, and the stock is surging on a claim that no independent benchmark has confirmed yet.

Published September 3, 2026, 2:48pm ET · 4 min read

Market Movers desk. Editor: David Moadel.

This post may contain links from our sponsors and affiliates, and Flywheel Publishing may receive compensation for actions taken through them.

A close-up shot of a hand holding a black smartphone showing the white Facebook 'f' logo on a vibrant blue screen. In the blurred background, the blue Meta logo, featuring an infinity symbol next to the word 'Meta', is visible against a light gray wall.
The Facebook app on a smartphone, set against the Meta logo, underscores Meta Platforms' ongoing relevance as a key player in social media and AI, as discussed in its Q2 2026 performance. © Fritz Jorgensen / iStock Editorial via Getty Images

The AI model race got a fresh salvo this morning, and it came from Menlo Park. Meta Platforms (NASDAQ:META | META Price Prediction) is claiming performance parity with the top private frontier labs, and buyers are showing up in size.

Meta Platforms stock is up 4% to $614.57, a jump that lands with extra weight given that Meta Platforms was down 10% year to date through the prior close. Meanwhile, the SPDR S&P 500 ETF Trust (NYSEARCA:SPY), which tracks the S&P 500, is up 1% to $773.75, so today’s enthusiasm sits inside this one name rather than the broader tape.

Alphabet (NASDAQ:GOOGL) is the other big-cap AI model rival investors judge Meta Platforms against, alongside privately held OpenAI and Anthropic. For sector-peer comparisons, Alphabet shares are up 2% to $342.61 and Reddit (NYSE:RDDT) stock is down 1% to $156.49.

META price target

Muse Spark 1.3 Powers the Move, according to Meta Platforms

Muse Spark 1.3 is Meta Platforms’ latest frontier push, an upgraded version of its flagship AI model that improves agentic and coding capabilities and is rolling out in Muse Code and the Meta Model API, with a max reasoning mode expected once additional safety testing wraps. The company stated, “Smarter and more practically useful, Muse Spark 1.3 advances our work toward personal superintelligence.” The rollout follows Muse Spark 1.1 and Muse Image, which Meta Platforms shipped earlier this year and wired into Facebook and Instagram recommendation systems.

The louder claim came from Alexandr Wang, Meta Platforms’ AI chief, who described Muse Spark 1.3 as the company’s biggest jump so far on model performance and said it now matches offerings from Anthropic and OpenAI. Developers can already access and pay for the model, with distribution across Meta Platforms’ social apps and the Meta AI assistant to follow. Management has also said its roadmap includes larger models and an eventual open weights release, keeping the door open on both closed and open ecosystems.

Claim vs. Confirmation

Today’s catalyst is a competitive assertion. No revenue attaches to a model release, so buyers are pricing the claim that Meta Platforms has closed the performance gap with the leading private labs. That claim comes from the company itself, and it arrives ahead of any independent benchmark result.

The context sharpens the stakes. Meta Platforms’ Q2 2026 print on July 29 delivered revenue of $60.80 billion, up 28% year over year, but diluted EPS of $6.18 missed the $7.22 estimate as legal and severance charges compressed operating margin. That miss snapped a six-quarter beat streak, left the stock in the penalty box, and set up today’s rally as a reset moment for the AI story.

Alphabet’s Q2 comparison is unflattering. Google Cloud revenue grew 82%, 90% of the Fortune 100 use Gemini Enterprise, and the Gemini App has 950 million monthly active users. Alphabet stock has ridden that dominance to a 60% one-year return, and Muse Spark 1.3 is what Meta Platforms hopes will start narrowing that competitive lead.

Meta Platforms has been laying groundwork all year. On the Q2 call, CEO Mark Zuckerberg said the company had seen “a 60% increase in the number of people interacting with the assistant each day” since it rebuilt Meta AI around the Muse Spark family, and business agents on WhatsApp and Messenger reached more than 1 million businesses per week. The Q1 2026 result had already delivered EPS of $10.44 against a $6.66 estimate, so the mid-year stumble looks more like a step in a longer buildout than a broken thesis.

What to Watch Next

Independent benchmark scores can decide whether today’s move ages well. Prediction markets currently give a probability of 0.48 that a Meta model scores 55% or higher on Humanity’s Last Exam this year, tapering to 0.255 at 60% and 0.125 at 65%. Shareholders can watch for third-party evaluations that either back or push back on Alexandr Wang’s parity claim.

Meta Platforms’ Connect event on September 23, 2026 is the next scheduled catalyst, and it’ll showcase how Muse Spark 1.3 lands in consumer surfaces, including new AI glasses developed with EssilorLuxottica. Traders could look for signs that independent labs and enterprise adopters echo Wang’s parity claim before extrapolating from one session’s move. Investors sizing new positions in META stock might stagger their exposure and keep their risk contained, because a company-authored parity claim can invite a rerating just as easily as it can invite skepticism if independent benchmarks disappoint.

Contact [email protected] for any questions or corrections.

David Moadel

David Moadel is financial writer specializing in stocks, ETFs, options, precious metals, and Bitcoin. David has written well over 1,000 articles for leading online publications, helping investors understand markets, income strategies, and risk.His work has appeared in The Motley Fool, InvestorPlace, U.S. News & World Report, TipRanks, ValueWalk, Benzinga, Market Realist, TalkMarkets, Finmasters, 24/7 Wall St., and others.With a master’s degree in education, David has taught at the elementary, high school, and college levels. That teaching background shapes his writing style: clear, educational, and practical. David has also built a loyal social-media audience by providing trustworthy financial content on YouTube, X/Twitter, and StockTwits.

All articles →