{"id":"sage-a1-5","slug":"sage-a1-5","name":"Sage-A1.5","orgId":"intern-science","family":"sage-a1-5","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://huggingface.co/InternScience/Sage-A1.5","hfRepo":null,"modalities":["text","image"],"aliases":["Sage-A1.5","sage-a1.5","sage a1.5","sage-a1-5","internscience/sage-a1.5","internscience sage-a1.5","intern science sage-a1.5"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-10-21","announced":"2026-10-21","blurb":"An open agentic model for long-horizon scientific work from InternScience at Shanghai AI Laboratory, fine-tuned from Qwen3.6-35B-A3B. It adds opt-in parallel subagents (Agentic Orchestration) and an opt-in in-reasoning self-review (Verification Mode), with native tool use and a 262K context. Apache 2.0.","api":"none","openWeights":"yes","params":{"text":"About 36B total, about 3B active (Qwen3.6-35B-A3B MoE)","source":"Hugging Face model card and safetensors metadata"},"facts":[{"label":"Base model","value":"Qwen3.6-35B-A3B"},{"label":"Context","value":"262,144 tokens"},{"label":"Inputs","value":"Text and images (includes a vision encoder; a text-only serving mode skips it)"},{"label":"Modes","value":"Agentic Orchestration (parallel subagents via a spawn_subagent tool) and Verification Mode (chat-template flag); both off by default"},{"label":"Predecessor","value":"Builds on InternScience's Agents-A1"},{"label":"Formats","value":"BF16, FP8 and GGUF (F16, Q8_0, Q4_K_M); Transformers, vLLM and SGLang"},{"label":"License","value":"Apache 2.0"},{"label":"Hugging Face","value":"Repo created Oct 15, 2026 (UTC); FP8 and GGUF builds and the ModelScope announcement on Oct 21"}],"claims":["Lab claim: 90.0 on PinchBench, 81.0 on FrontierScience-Olympiad and 48.3 on FrontierScience-Research, ahead of Kimi-K3, GLM-5.3 and GPT-5.6-sol in InternScience's comparison.","Lab claim: overall best in its comparison on XBench-DS-2510 (92.0), IFBench (81.6), CMT (80.0) and OmniDocBench (91.5), and 95.1 on GAIA."],"sources":[{"kind":"repo","title":"Sage-A1.5 on Hugging Face","url":"https://huggingface.co/InternScience/Sage-A1.5","date":"2026-10-21"},{"kind":"lab","title":"Sage-A1.5 project page","url":"https://alpha-innovator.github.io/Sage-A1.5/","date":"2026-10-21"},{"kind":"press","title":"ModelScope announcement on X","url":"https://x.com/ModelScope2022/status/2112816255375745132","date":"2026-10-21"}],"firstBenchmark":null},"organization":{"id":"intern-science","name":"InternScience (Shanghai AI Laboratory)","slug":"intern-science","country":null,"url":"https://huggingface.co/InternScience","color":"#11110F","mark":"I"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Oct 21, 2026; no independent results yet.","lead":"As of Oct 22, 2026, Sage-A1.5 is pending benchmarks on isitsota: announced Oct 21, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}