Muse Spark 1.2 (xHigh) Climbs to #11 on Vision Arena Leaderboard
Meta's Muse Spark 1.2 (xHigh) multimodal model has reached the #11 spot on the Vision Arena leaderboard with 1,290 points. This marks a notable improvement from its predecessor, Muse Spark 1.1, which was ranked at #20. This update demonstrates rapid progress in Meta's multimodal reasoning capabilities, particularly in visual tasks. It highlights the competitive nature of the vision-language model landscape as developers continuously optimize performance. The version 1.2 upgrade showed its largest category gains in HTML (jumping from #24 to #8), Gaming (#23 to #13), Frontend (#19 to #13), and Data & Analytics (#13 to #8).
## BACKGROUND
Muse Spark is a multimodal reasoning model developed by Meta Superintelligence Labs (MSL) designed for tool-use, visual chain of thought, and multi-agent orchestration. The Vision Arena, hosted by LMSYS, is a crowdsourced benchmark platform where users compare and vote on the outputs of different vision-language models to establish a global leaderboard.