MODEL PROFILE · 模型档案

Ling 3.1 Flash 1019

InclusionAI · ling-3-1-flash-1019

🤗 Hugging Face    |   🤖 ModelScope    |   🐙 OpenRouter    We introduce **Ling-3.1-flash-1019**, an upgraded version of Ling-3.1-flash, with **560B** total parameters, **25B** active parameters per token, and a **1M-token context window**. With further post-training optimizations, **Ling-3.1-flash-1019** demonstrates significant improvements in reasoning and real-world agentic capabilities. Its **AA-Briefcase** score rises from approximately 1400 to **1524 Elo**, while performance on **AutomationBench-AA** increases from 62% to **68%**. The model also makes gains on scientific benchmarks, including **CritPt** (18% → **26%**) and **SciCode** (54% → **57%**). Overall, it scores **44** on the **Artificial Analysis Intelligence Index v4.3.2**, up from 41 for its predecessor. These gains highlight the model's improved capabilities in complex reasoning, multi-step task execution, and agentic workflows across diverse real-world scenarios. The table and diagram below summarize the model’s specifications and architecture.

译文排队中,当前展示来源原文。

文本生成license:mit
上下文窗口1,048,576 Token
输入 → 输出文本 → 文本
模型权重开放权重
档案更新时间2026/11/16 16:30:11
OVERVIEW

模型介绍

🤗 Hugging Face    |   🤖 ModelScope    |   🐙 OpenRouter    We introduce **Ling-3.1-flash-1019**, an upgraded version of Ling-3.1-flash, with **560B** total parameters, **25B** active parameters per token, and a **1M-token context window**. With further post-training optimizations, **Ling-3.1-flash-1019** demonstrates significant improvements in reasoning and real-world agentic capabilities. Its **AA-Briefcase** score rises from approximately 1400 to **1524 Elo**, while performance on **AutomationBench-AA** increases from 62% to **68%**. The model also makes gains on scientific benchmarks, including **CritPt** (18% → **26%**) and **SciCode** (54% → **57%**). Overall, it scores **44** on the **Artificial Analysis Intelligence Index v4.3.2**, up from 41 for its predecessor. These gains highlight the model's improved capabilities in complex reasoning, multi-step task execution, and agentic workflows across diverse real-world scenarios. The table and diagram below summarize the model’s specifications and architecture.

下方摘要由共用资料处理模型依据来源整理,非本站实测;选型结论请结合原始资料和实际业务验证。

主要特点

  • 支持1M-token超长上下文窗口(1048576 tokens)
  • 采用稀疏激活架构,总参数量560B,每token激活25B参数
  • 在推理、自动化任务与科学能力方面较前代显著提升,AA-Briefcase Elo分达1524,AutomationBench-AA达68%

局限与注意事项

暂无足够来源支持具体缺点结论。请重点验证实际任务表现、时延、数据政策、许可与版本生命周期。

PRICING & COST

来源报价与费用分析

同一模型在不同渠道、地区和配置下可能采用不同价格。下表独立保留各渠道报价,不将渠道价格视为厂商官方统一价格。

来源 / 渠道输入输出计费单位采集时间
Artificial Analysis · Ling 3.1 Flash 1019 ↗USD 0.3USD 0.9每百万 Token2026/11/2 06:30:46

部分渠道提供缓存优惠。缓存读取、写入及额外费用详见对应报价来源。

按月估算基础 Token 费用

—

预估不包含缓存优惠、阶梯定价、多模态附加费用、网络检索、税费及推理额外输出。

INDEPENDENT BENCHMARKS

公开评测与运行配置

Terminal-Bench 4.00.33Ling 3.1 Flash 1019Artificial Analysis ↗采集于 2026/11/2 06:30:46
AA 智能指数29.47Ling 3.1 Flash 1019Artificial Analysis ↗采集于 2026/11/2 06:30:46
查看完整已收录榜单 →

常见问题

Ling-3.1-flash-1019的上下文长度是多少?

上下文窗口为1M token(即1048576 tokens)。

该模型的参数规模结构是怎样的?

总参数量为560B,采用稀疏激活机制,每token仅激活25B参数。

它在哪些权威基准上取得了性能提升?

在AA-Briefcase(Elo 1524)、AutomationBench-AA(68%)、CritPt(26%)、SciCode(57%)及Artificial Analysis Intelligence Index v4.3.2(44分)等基准上均较前代提升。

开放权重是否代表可以免费商用?

不一定。应核对具体版本许可证、商用门槛和使用政策。自部署还会产生算力和维护成本。

这些价格是实时结算价格吗?

不是实时结算价格,而是带采集时间的来源报价快照。请到对应渠道核对当前版本、区域、折扣和计费方式。

资料来源与可追溯记录

许可证:MIT。模型介绍与第三方评测均以原来源的具体版本和方法为准。