--- id: "2026-10-14-ling-3-1-flash-open-weights" url: "https://postcutoff.com/e/2026-10-14-ling-3-1-flash-open-weights/" as_of: "2026-10-14T22:23:00+02:00" date: "2026-10-14" date_precision: day category: open-source importance: 3 confidence: high status: [Confirmed] sources: 5 editor: human editor human_review: null version: null --- As of: 2026-10-14 22:23 CEST. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/e/2026-10-14-ling-3-1-flash-open-weights/ # inclusionAI releases Ling 3.1 Flash open weights Full title: inclusionAI releases Ling 3.1 Flash open weights: a 560B-parameter MIT-licensed MoE with a 1M-token context Two weeks after its API-only free trial, Ant Group's inclusionAI published the weights of Ling 3.1 Flash on Hugging Face under the MIT license (repo created Oct 13, files finished Oct 14, 2026). It is a 560B-total / 25B-active hybrid-linear mixture-of-experts reasoning model with a 1M-token context, which makes it one of the largest openly licensed models from a Chinese lab. Benchmark results are vendor-reported and shown only as images. ## Key facts - Hugging Face: inclusionAI/Ling-3.1-flash, license MIT, 54 safetensors shards; repo created 2026-10-13, last modified 2026-10-14 (HF API) - Architecture (model card): hybrid-linear MoE, 560B total / 25B active, 72 layers = 63 KDA (linear attention) + 9 gated MLA layers (7:1), 512 routed experts + 1 shared, 8 active, vocabulary 157,184 - Context window: 1M tokens (the trial was capped at 256K) - Model card names GDPVal-AA V2.1, HealthBench Professional, FrontierSWE, Terminal-Bench 4.0, SWE-bench Pro, MCP-Atlas and SkillsBench as strong results; scores are only in images, not text - Serving: day-0 SGLang image (8x H200 TP8+EP8, or 4x GB300) with MTP speculative decoding; vLLM branch also supported - Also introduces LingCode, a long-horizon coding harness built on the open-source Pi agent; inclusionAI is 'considering open-sourcing it in the future' - A training-content summary (TDS) PDF is published in the inclusionAI/AI-Transparency repo ## What happened Ling 3.1 Flash launched around Sept 29–30, 2026 as a two-week free trial on API gateways, with a promise to open-source it afterwards. The trial ended around Oct 13. On Oct 13–14 inclusionAI published the full weights on Hugging Face (`inclusionAI/Ling-3.1-flash`, 54 safetensors shards, MIT license), with a model card giving the architecture: a hybrid of 63 Kimi-style delta attention (KDA) linear layers and 9 gated MLA layers, 512 routed experts with 8 active, 25B active parameters out of 560B, and a 1M-token context. The card claims strong results on agentic and office benchmarks (GDPVal-AA V2.1, Terminal-Bench 4.0, FrontierSWE, SWE-bench Pro and others) but gives the numbers only as images, so they are not quoted here. It also announces LingCode, a coding-agent harness built on the open-source Pi agent, which is not yet released. No independent evaluations had appeared by the evening of Oct 14. ## Why it matters inclusionAI kept its promise on time: the API-first, weights-later pattern was followed through, unlike StepFun's Step 5 Preview, whose Oct 15 weights were still not public on Oct 14. A 560B model under the permissive MIT license, with a 1M-token context and a mostly linear-attention design, gives open-model users a large long-context option. Its claimed scores are unverified until outside tests appear. ## Your AI and this story - GPT-6 Astra (training cutoff April 2026): 167 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 106 days after its cutoff - Gemini 4 Argon (training cutoff March 2026): 197 days after its cutoff - Grok 4.7 (training cutoff May 2026): 136 days after its cutoff ## Sources 1. [Hugging Face: inclusionAI/Ling-3.1-flash (model card)](https://huggingface.co/inclusionAI/Ling-3.1-flash) (huggingface.co, code) 2. [Hugging Face API: repo metadata (created/modified dates, license, files)](https://huggingface.co/api/models/inclusionAI/Ling-3.1-flash) (huggingface.co, docs) 3. [SGLang cookbook: Ling-3.1-flash](https://docs.sglang.io/cookbook/autoregressive/InclusionAI/Ling-3.1-flash) (docs.sglang.io, docs) 4. [inclusionAI: Ling-3.1 training-content summary (PDF)](https://huggingface.co/inclusionAI/AI-Transparency/resolve/main/Ling-3.1_TDS-Summary.pdf) (huggingface.co, docs) 5. [TechNode (Sept 30): Ant Group launches Ling-3.1-flash, open source promised after trial](https://technode.com/2026/09/30/ant-group-launches-ling-3-1-flash-with-560-billion-parameters/) (technode.com, press) ## Changes - 2026-10-14 (filed): Created (HF repo and model card read; outcome of upcoming item [inclusionAI releases Ling 3.1 Flash open weights](https://postcutoff.com/e/2026-10-14-ling-3-1-flash-open-weights/)) ## Related - 2026-09-29: [Ant Group's inclusionAI launches Ling 3.1 Flash, a 560B-parameter MoE, as a free trial with open weights promised](https://postcutoff.com/e/2026-09-29-ant-inclusionai-ling-3-1-flash/index.md) - 2026-09-20: [StepFun launches Step 5 Preview, a 600B-parameter MoE agent model with 1M context](https://postcutoff.com/e/2026-09-20-stepfun-step-5-preview/index.md) - Models: [Ling 3.1 Flash](https://postcutoff.com/m/ling-3-1-flash/) (`ling-3-1-flash`)