{"id":"sage-s2-1-flash","slug":"sage-s2-1-flash","name":"Sage S2.1 Flash","orgId":"intern-large-models","family":"sage-s2-1-flash","releaseDate":null,"availability":"ga","openWeights":true,"license":"MIT (official model card)","paramsTotal":321323031390,"paramsSource":"tracked","contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://internlm.github.io/Sage-S2.1-Flash/","hfRepo":null,"modalities":["text","image"],"aliases":["Sage S2.1 Flash","Sage-S2.1-Flash","sage s2.1 flash","sage-s2.1-flash","sage-s2-1-flash","internlm/Sage-S2.1-Flash","internlm/sage-s2.1-flash","书生 S2.1 Flash","Shanghai AI Laboratory Sage S2.1 Flash"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-11-16","announced":"2026-11-16","blurb":"Shanghai Artificial Intelligence Laboratory's scientific multimodal model (书生 S2.1 Flash). Official materials describe a 320B-A18B mixture-of-experts that uses the GLM-5.3-Flash architecture. Through Sage SciPerception it takes more than ten kinds of scientific input — including molecular graphs, crystal structures, protein sequences and 3D structures, DNA and RNA sequences, spectra, mass spectra, X-ray diffraction, electrocardiograms, electroencephalograms, and light curves — alongside natural-language instructions, and combines scientific data understanding with retrieval, programming, and tool use. Announced November 16, 2026 (official website and InternLM GitHub repository created 2026-11-16T03:34:03Z, 10:34 PM ET November 15; team posts later the same calendar day in ET). Open MIT-licensed weights, ungated on Hugging Face at internlm/Sage-S2.1-Flash and on ModelScope, with an FP8 build. No documented public self-serve API model id was confirmed.","api":"none","openWeights":"yes","params":{"text":"320B-A18B MoE (official deployment guide); official website says 320-billion-parameter MoE. Hugging Face safetensors metadata totals 321,323,031,390 parameters","source":"Official Sage-S2.1-Flash deployment guide, official website, and Hugging Face safetensors metadata","total":321323031390},"facts":[{"label":"Developer","value":"Shanghai Artificial Intelligence Laboratory (official model card and website)"},{"label":"Architecture","value":"Official deployment guide: 320B-A18B multimodal MoE in bfloat16 (about 645 GB), using the GLM-5.3-Flash architecture. Official config.json: Glm5NextForConditionalGeneration / glm5_next, 45 layers, 288 routed experts plus 1 shared, 8 experts per token"},{"label":"Context","value":"1,048,576 tokens (max_position_embeddings in official config.json)"},{"label":"Input","value":"Text and images, plus scientific inputs via Sage SciPerception: molecular graphs, crystal structures, protein sequences and 3D structures, DNA and RNA sequences, spectra, mass spectra, X-ray diffraction, electrocardiograms, electroencephalograms, and light curves (official website and model card)"},{"label":"License","value":"MIT (official model card)"},{"label":"Weights","value":"Ungated internlm/Sage-S2.1-Flash (BF16) and internlm/Sage-S2.1-Flash-FP8 on Hugging Face; mirrored on ModelScope under Shanghai_AI_Laboratory"},{"label":"Hosted access","value":"Official website links Try the API to https://discovery.intern-ai.org.cn/ (Intern InkStone). A documented public self-serve API model id was not confirmed without an account"},{"label":"Related","value":"Separate from Sage-A1.5 (InternScience), Sage2 and Sage 1.3.0 (Levanto Labs), Atria Dawn, and GLM-5.3-Flash"}],"claims":["Lab claim: self-reported official model-card scores of MLE-bench Lite 90.3, OfficeQA Pro 72.2, SpreadsheetBench 2 37.1, Terminal-Bench 2.1 87.9, DeepSWE 67.9, and SWE-bench Pro 60.6."],"sources":[{"kind":"lab","title":"Sage S2.1 Flash official website","url":"https://internlm.github.io/Sage-S2.1-Flash/","date":"2026-11-16"},{"kind":"repo","title":"internlm/Sage-S2.1-Flash on Hugging Face","url":"https://huggingface.co/internlm/Sage-S2.1-Flash","date":"2026-11-16"},{"kind":"lab","title":"Sage S2.1 Flash deployment guide","url":"https://huggingface.co/internlm/Sage-S2.1-Flash/blob/main/deployment_guide.md","date":"2026-11-16"},{"kind":"repo","title":"internlm/Sage-S2.1-Flash-FP8 on Hugging Face","url":"https://huggingface.co/internlm/Sage-S2.1-Flash-FP8"},{"kind":"repo","title":"Shanghai_AI_Laboratory/Sage-S2.1-Flash on ModelScope","url":"https://www.modelscope.cn/models/Shanghai_AI_Laboratory/Sage-S2.1-Flash"},{"kind":"repo","title":"InternLM/Sage-S2.1-Flash on GitHub","url":"https://github.com/InternLM/Sage-S2.1-Flash","date":"2026-11-16"},{"kind":"lab","title":"Team post on X (quoted announcement)","url":"https://x.com/i/status/2122319410623029578","date":"2026-11-16"},{"kind":"lab","title":"Team post on X","url":"https://x.com/wanghan_xu/status/2122378890689851575","date":"2026-11-16"}],"firstBenchmark":null},"organization":{"id":"intern-large-models","name":"Intern Large Models","slug":"intern-large-models","country":null,"url":"https://huggingface.co/internlm","color":"#A39E8F","mark":"IN"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Nov 16, 2026; no independent results yet.","lead":"As of Nov 22, 2026, Sage S2.1 Flash is pending benchmarks on isitsota: announced Nov 16, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}