{"id":"ling-3-2-flash","slug":"ling-3-2-flash","name":"Ling-3.2-flash","orgId":"inclusionai","family":"ling-3-2-flash","releaseDate":null,"availability":"ga","openWeights":false,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://developer.ant-ling.com/en/docs/models/ling/","hfRepo":null,"modalities":["text","image","video"],"aliases":["Ling-3.2-flash","ling-3-2-flash","ling-3.2-flash","ling 3.2 flash","ant ling 3.2 flash","inclusionai/ling-3.2-flash"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-05","announced":"2026-11-05","blurb":"Ant Ling's native multimodal MoE model in the Ling series: 560B total parameters with about 25B activated per token, a 1M-token context window, and native image, text and video input. Positioned for complex visual understanding, multimodal long-horizon agents, 3D understanding and interaction, and medical multimodal work. Available through the Ant Ling API (OpenAI-compatible model id Ling-3.2-flash) and on Computrix demos. Open weights were not on inclusionAI Hugging Face at tracking time.","api":"live","openWeights":"no","params":{"text":"560B total / 25B active","source":"Ant Ling developer docs and AntLingAGI X post"},"facts":[{"label":"Architecture","value":"Mixture-of-Experts; native multimodal (vision fused in the architecture, not an external vision module)"},{"label":"Context","value":"Up to 1M tokens"},{"label":"Modalities","value":"Image, text and video input"},{"label":"Focus areas","value":"Complex visual understanding, multimodal long-horizon Agent, 3D understanding and interaction, medical multimodal"},{"label":"API","value":"Ant Ling developer platform; OpenAI-compatible model id Ling-3.2-flash"},{"label":"Demo","value":"Computrix (Ant Group) per the lab's launch materials"},{"label":"Weights","value":"Not published on inclusionAI Hugging Face as of 2026-11-05 (Ling-3.1-flash and Ling-3.1-flash-1019 are; this checkpoint is not)"},{"label":"Related","value":"Separate from Ling-3.1-flash, Ling-3.1-flash-1019 and Ling-3.1-flash-base"}],"claims":["Lab claim: on OmniDocBench1.5, 92.21 vs 90.56 for Opus 5 (max).","Lab claim: on ZEROBench (Sub), 43.71 vs 39.52 for Opus 5 (max).","Lab claim: on Vision2Web, 85.86 vs 80.20 for Opus 5 (max)."],"sources":[{"kind":"lab","title":"Ling model docs (Ant Ling developer platform)","url":"https://developer.ant-ling.com/en/docs/models/ling/","date":"2026-11-05"},{"kind":"lab","title":"AntLingAGI on X: Ling-3.2-flash","url":"https://x.com/AntLingAGI/status/2118428246664761379","date":"2026-11-05"}],"firstBenchmark":null},"organization":{"id":"inclusionai","name":"Ant Ling (InclusionAI, Ant Group)","slug":"inclusionai","country":null,"url":null,"color":"#11110F","mark":"A"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Nov 5, 2026; no independent results yet.","lead":"As of Nov 22, 2026, Ling-3.2-flash is pending benchmarks on isitsota: announced Nov 5, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}