--- format: "aidr-story-markdown/v1" id: "a3ea997271a7967aacb445fd65393e4f247a80e9c76ed90ae2689a713ddcc29c" canonical_url: "https://aidr.today/a3ea9972?lang=en" title: "Ant Ling’s New Ling-3.2-flash Model Joins Top 10 of DesignArena Leaderboard" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-11-05T20:09:17.000Z" category: "Models" topics: ["llm"] source_urls: ["https://huggingnews.com/ai/ant-lings-new-ling-32-flash-model-joins-top-10-of-designarena-leaderboar-dd43f82e","https://marketbrief.now/ai/ant-lings-new-ling-32-flash-model-joins-top-10-of-designarena-leaderboar-dd43f82e"] summary: "Ant Ling's new native AI model, Ling-3.2-flash, consists of 560B total parameters and 25B active parameters and has joined the top 10 of the DesignArena leaderboard with an Elo score of 1238. The model supports medical intelligence and complex visual understanding, outscored Opus 5 (max) in release evaluations on the ZEROBench(Sub), Vision2Web, and OmniDocBench1.5 benchmarks, and is priced at $0.30 for input and $0.90 for output per 1M tokens, noted as the lowest rate for this level of performance." --- # Ant Ling’s New Ling\-3\.2\-flash Model Joins Top 10 of DesignArena Leaderboard > [Open the canonical story]() **Published:** 2026-11-05T20:09:17.000Z **Category:** Models **Topics:** llm ## Summary Ant Ling's new native AI model, Ling\-3\.2\-flash, consists of 560B total parameters and 25B active parameters and has joined the top 10 of the DesignArena leaderboard with an Elo score of 1238\. The model supports medical intelligence and complex visual understanding, outscored Opus 5 \(max\) in release evaluations on the ZEROBench\(Sub\), Vision2Web, and OmniDocBench1\.5 benchmarks, and is priced at $0\.30 for input and $0\.90 for output per 1M tokens, noted as the lowest rate for this level of performance\. ## Sources - [Story source]() - [Story source]()