{"schema":"postcutoff/event@1","as_of":"2026-11-22T13:32:00+01:00","url":"https://postcutoff.com/e/2026-11-15-shanghai-ai-lab-sage-s2-1-flash-open-weights/","md":"https://postcutoff.com/e/2026-11-15-shanghai-ai-lab-sage-s2-1-flash-open-weights/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-11-15-shanghai-ai-lab-sage-s2-1-flash-open-weights","date":"2026-11-15","date_precision":"day","short_title":"Sage S2.1-Flash open weights","deck":null,"takeaway":"On 15 November 2026 the InternLM organisation of the Shanghai AI Laboratory uploaded Sage S2.1-Flash to Hugging Face under the MIT licence, without a press release.","category":"open-source","category_label":"Open source","importance":2,"confidence":"medium","status":{"key":"partly","labels":["Partly confirmed"]},"sources":[{"n":1,"title":"Hugging Face: internlm/Sage-S2.1-Flash","url":"https://huggingface.co/internlm/Sage-S2.1-Flash","type":"code","group":"primary","domain":"huggingface.co"},{"n":2,"title":"AI Weekly: InternLM's 321B Sage S2.1 Flash leads MLE-bench Lite at 90.3","url":"https://aiweekly.co/alerts/internlms-321b-sage-s21-flash-leads-mle-bench-lite-at-903","type":"press","group":"press","domain":"aiweekly.co"},{"n":3,"title":"OrcaRouter: Sage S2.1 Flash's quiet release","url":"https://www.orcarouter.ai/blog/sage-s2-1-flash-quiet-release","type":"discussion","group":"reactions","domain":"orcarouter.ai"}],"official":1,"filed":"2026-11-18","updated":"2026-11-18","orgs":["Shanghai AI Laboratory","InternLM"],"title":"Shanghai AI Lab quietly posts Sage S2.1-Flash, a ~320B open-weight scientific multimodal MoE model (MIT licence)","summary":"On 15 November 2026 the InternLM organisation of the Shanghai AI Laboratory uploaded Sage S2.1-Flash to Hugging Face under the MIT licence, without a press release. OrcaRouter describes it as a 320B-total, 18B-active MoE in bfloat16 (~645 GB); AI Weekly reports it handles 10+ scientific input types (molecular graphs, crystal structures, protein sequences and structures) and leads MLE-bench Lite at 90.3 ahead of DeepSeek V4.1 Flash (89.2). The benchmark claims are single-source and unverified here.","key_facts":["Hugging Face repo internlm/Sage-S2.1-Flash created 2026-11-15 02:34 UTC, last modified Nov 16; model card body empty at the time OrcaRouter checked; architecture tag 'glm5_next' (HF API, read Nov 18: 678 downloads, 125 likes)","Reported size: 320B total / 18B active (OrcaRouter) or 321B (AI Weekly); BF16 weights plus an FP8 variant (AI Weekly)","Claimed: MLE-bench Lite 90.3, first, vs DeepSeek-V4.1-Flash-1017 89.2 (AI Weekly; results in a separate GitHub repo per OrcaRouter)"],"key_numbers":[],"tags":["open-weights","science","multimodal","moe","china","internlm"],"science":null,"body_md":"## What happened\n\nA large Chinese open-weight science model appeared without announcement; found through a web search on 18 Nov, not through the sweep (no lab feed or Techmeme story).","disputed":[],"related":[],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-11-18","type":"filed","text":"Created. Benchmark claims single-source"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-11-18","run":null,"sources_read":null,"updated":"2026-11-18","human_review":null,"version":{"date":"2026-11-18"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":168,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":138,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":229,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":168,"in_training_data":false}],"short_url":null}