Many users praise Meta's Muse Spark 1.1 for its video-based multimodal reasoning that automates real-world tasks like marketplace listings, while a few criticize output quality and competitiveness.
Based on 19 visible X reactions from 31 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@alexandr_wang this is how it writes, can you read this? constantly again and again full of broken text/math. its not a good model sir. https://x.com/AlonLaleza30319/status/2075889191209083171/photo/1
@rohanpaul_ai the multimodal reasoning capabilities of muse spark 1.1, allowing it to extract data from video and operate the browser autonomously, is a massive step forward for ai agents.
@rohanpaul_ai it's impressive how the model doesn't just recognize the object, but follows through with active steps like uploading images and publishing listings.
@alexandr_wang i never thought in my life i would commend Meta for a great job but for Muse I will say it; bravo https://x.com/alexandr_wang/status/2075652012608467385?s=20
@alexandr_wang Video input for real world tasks is the next frontier and Meta gets it.
@alexandr_wang You immature little boy.
Meta's Muse Spark 1.1 's multimodal reasoning ability. Here it took videos from mobile, reaons to extract photos & product details then operates the browser with computer use actions to fill Marketplace forms, upload images and publish listings. https://x.com/AIatMeta/status/2075221102172016989/video/1
muse spark is able to do end-to-end tasks based on short video instructions https://twitter.com/aiatmeta/status/2075221102172016989
Many users praise Meta's Muse Spark 1.1 for its video-based multimodal reasoning that automates real-world tasks like marketplace listings, while a few criticize output quality and competitiveness.
Based on 19 visible X reactions from 31 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@alexandr_wang You immature little boy.