ainewsblitz.com

Breaking

Meta's Muse Spark 1.2 Joins Agent Arena Leaderboard, Posts +2.1% Net Improvement Behind the Leaders

  • AI Agents
  • Foundation Models
  • Software Dev & Coding

Meta's newest coding-focused model has made its debut on one of the industry's toughest agentic benchmarks, landing squarely in the middle of the pack. Arena added Muse Spark 1.2, running at its highest "xHigh" reasoning setting, to Agent Arena this week, where the model posted a net improvement of +2.1% (2.06%, with a margin of ±1.14 percentage points, in Arena's own published figures) — a positive score, but one that ranks it 25th out of 51 models on the leaderboard as of Arena's most recent snapshot, taken August 19.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,555 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year