Ant Group's Bailing open-sources Ling-3.1-flash-1019 for personal agents
On October 28, 2026, Ant Group's Bailing open-sourced Ling-3.1-flash-1019, an updated model for personal-agent tasks, along with its base model Ling-3.1-flash-base. The new version scores 44 on the Artificial Analysis Intelligence Index v4.3.2, up 3 points from the previous version, and shows improvements across benchmarks including AA-Briefcase, GDPval-AA, AutomationBench-AA, CritPt, and SciCode. The release is linked to the Lingguang 2.0 upgrade, which positions the assistant as a proactive personal AI that can manage long-running tasks.
Coverage timeline
量子位量子位
Personal Agent走向真实任务，蚂蚁百灵开源 Ling-3.1-flash-1019 量子位的朋友们 2026-10-28 12:04:48 来源： 量子位 10月28 日，蚂蚁集团旗下百灵大模型正式开源 Ling-3.1-flash-1019。 10月28 日，蚂蚁集团旗下百灵大模型正式开源 Ling-3.1-flash-1019。新版本在 Ling-3.1-flash 基础上进一步训练优化，在智能指数Artificial Analysis Intelligence Index v4.3.2 中获得44分，较上一版本提升3分。与此同时，百灵同步开源其基础模型 Ling-3.1-flash-base，为长程任务所需的知识、推理和计算效率打好基础。 据了解，Ling-3.1-flash-1019 版本面向 Personal Agent 场景，进一步提升了模型处理复杂工作流和多步骤任务的表现。从测评具体来看，新版本在复杂知识工作AA-Briefcase 中的得分从约1400 Elo提升至1524 Elo，GDPval-AA从约1618 Elo提升至1644 Elo；流程自动化AutomationBench-AA得分从62%提升至68%；科学问题求解CritPt得分从18%提升至26%，SciCode得分从54%提升至57%。 记者注意到，百灵本次模型迭代开源与灵光 2.0 息息相关。就在 10 月 27日，灵光 2.0 升级为「能管事的个人AI助理」，依托跨会话记忆、长程任务调度与独立云主机等能力，灵光可以在用户授权范围内更主动地跟进任务、持续管理生活中的复杂事务，即使用户退出对话，相关任务也能继续运行。这背后对模型能力提出了新要求。 从回答问题到持续办事，模型需要哪些新能力？ 与即时问答不同，个人智能体需要围绕一个目标持续工作，这既需要产品具备记忆和任务调度机制，也需要底层模型能够理解目标、使用工具、处理新信息并推进任务。 灵光2.0，呈现了这类能力在个人生活中的应用。 例如，规划一趟旅行，不只是生成一份行程攻略，还需要结合出发时间跟进天气、预约和行前准备；面对双11购物需求，则要按照预算、规格和个人偏好筛选商品、整理清单，并持续关注价格变化，在符合约定条件时提醒用户。 围绕这些需求，百灵持续通过模型训练和应用反馈，探索 Personal Agent 所需的
Related stories
Ant Group open-sources Ling-3.1-flash, a 560B-parameter MoE model with 1M context
Ant Group's Ling-3.1-flash, a 560B-parameter mixture-of-experts model with 25B active parameters and a 1M-token context window, was released open-source on October 15. The model targets long-horizon complex task execution, using a hybrid linear attention design with KDA layers and Gated MLA to manage computational costs on long sequences.
1 reports · 1 sources
MiniMax unveils M3.1 Flash, claiming top agent scores and low cost
MiniMax released M3.1 Flash, the official version of the previously anonymous Space Bunny model, claiming a generational leap in agentic tasks, coding, computer use, and cybersecurity. The model scores 29.6% on Agents' Last Exam (ALE), the top for Chinese models, with pricing at 1 yuan per million input tokens and 4 yuan per million output tokens. Weights are due October 23, with day-0 support from vLLM on Blackwell, Rubin, and AMD MI350 GPUs.
4 reports · 4 sources
Anthropic donates Agent Skills standard to AAIF, 70+ platforms integrate
Anthropic announced on October 22, 2026, that it is donating the Agent Skills open standard to the Agentic AI Foundation (AAIF) under the Linux Foundation, which it co-founded with Block and OpenAI. More than 70 agent platforms, including ChatGPT, Cursor, GitHub Copilot, and VS Code, have integrated Agent Skills since the standard was published in December 2025. The move follows Anthropic's earlier donation of the Model Context Protocol (MCP) to AAIF in 2025.
Liquid AI releases LFM2.5-24B-A2B, an open-weight 24B reasoning model with 2B active parameters
Liquid AI announced LFM2.5-24B-A2B, an open-weight 24B reasoning model with 2B active parameters, designed for fast, private on-device agents. The company claims it outperforms Qwen3.6-35B-A3B on the LiveClaw benchmark.
2 reports · 2 sources287 points · 80 comments
-oOdUn61bDRSykk4H951SoDOHzqoOe1.png)
News updates every hour. Subscribe in your RSS reader to get new stories automatically.
Subscribe