Radio
Now Playing
Quickyla Radio โ€” Click to play
Open โ†’
3 min left
Back to News

Meta releases Muse Spark 1.3 with strong performance, limited access for developers.

Meta has released Muse Spark 1.3, an AI model that offers improved performance for coding tasks, but its best features are currently not widely accessible due to ongoing safety testing. This limitatiโ€ฆ

Meta says Muse Spark 1.3 has frontier performance โ€” but its best results come from a model developers canโ€™t broadly use yet
VentureBeat โ€” 3 September 2026
Text:
1 0 0

Meta has launched Muse Spark 1.3, its latest AI model, which promises enhanced performance and speed compared to its predecessor. The company revealed the update yesterday, touting it as a significant advancement in coding and agentic tasks. Mark Zuckerberg, Meta's co-founder and CEO, described the new model as having "frontier performance almost too cheap to meter" on social media platform X.

The release comes at a time when competition in the AI space is intensifying. Developers are constantly seeking tools that can boost productivity and efficiency. Muse Spark 1.3 is designed to meet that demand, showcasing notable improvements over the previous version, Muse Spark 1.2. According to Meta, the new model excels particularly in long-running tasks, making it a valuable asset for developers who rely on robust AI capabilities.

Despite the excitement surrounding Muse Spark 1.3, there is a caveat. The most impressive results stem from a max reasoning configuration that is not yet broadly available. Meta has indicated that this particular version is undergoing additional safety testing and will be released "shortly." Meanwhile, a third-party benchmarking firm, Artificial Analysis, has evaluated the max configuration under a limited preview but has not yet associated it with any API provider.

As the AI landscape evolves, the release of Muse Spark 1.3 could have significant implications for developers and businesses. The model's strong price-performance ratio positions it competitively in the market, but the limited availability of its best features may hinder widespread adoption initially. Observers will be watching closely to see how quickly Meta can roll out the max reasoning configuration and what impact it will have on the AI development ecosystem.

Read Full Story at VentureBeat โ†’
Advertisement
"frontier performance almost too cheap to meter"
โ€” VentureBeat
React:
Sources
Sponsored

More to Read

Tokyo's Sushi Bus combines conveyor belt dining with open-aโ€ฆ
๐Ÿ’ป Technology
Tokyo's Sushi Bus combines conveyor belt dining with open-air sightseeing
Engadget ยท 13 days ago
The wide-screen phone trend is expanding beyond foldables
๐Ÿ’ป Technology
The wide-screen phone trend is expanding beyond foldables
Android Authority ยท 14 days ago
LinkedIn says its AI slop button is working
๐Ÿ’ป Technology
LinkedIn says its AI slop button is working
Engadget ยท 13 days ago
Nigeria's jet fuel conundrum: Scarcity at home, abundance aโ€ฆ
๐ŸŒ World News
Nigeria's jet fuel conundrum: Scarcity at home, abundance abroad
DW World ยท 9 days ago
Is Sudanโ€™s battlefield shaping the terms of its next politiโ€ฆ
๐ŸŒ World News
Is Sudanโ€™s battlefield shaping the terms of its next political phase?
Al Jazeera ยท 9 days ago
Firms scramble for battery power in Spain and Portugal
๐Ÿ’ฐ Business
Firms scramble for battery power in Spain and Portugal
BBC Business ยท 9 days ago
Full view