DATA CUT 2026-07-27 116 活跃模型 680 A 类案例

模型档案

Seed1.5 VL

Seed1.5 VL 已有可核验 A 类案例,可优先进入选型对比。

已有真实案例

档案摘要 / Decision snapshot

先判断能不能用于选型。

证据完整度 100%
公开状态 已有真实案例

可进入对比

案例覆盖 8 条 A 类

有真实任务证据

主要适用 作为完整模型索引和厂商路线追踪入口

来自模型卡

谨慎场景 缺少真实案例时不要包装成推荐

边界优先

30 秒结论

可进入选型对比,但仍以案例证据为准。

来自厂商页骨架的补充模型;缺失字段以“暂无数据 / 官方未披露”展示。

这页有来源、有风险说明、有公开可核验 A 类案例。

数据可信度 / 证据强度

已有真实案例

A 类案例

8

AA 评分

暂无数据

官方来源

2 个补充入口

案例来源

8 条可核验

来自厂商页/模型卡骨架 + 2 个厂商/官方/模型家族证据入口;暂无 Artificial Analysis 行。

Evidence Distribution

案例证据分布

Seed1.5 VL 当前关联 8 条 A 类案例;这里按任务、来源和复核时间观察证据结构。

A CASES 8 8 selected

Evidence Snapshot

证据快照覆盖

Seed1.5 VL 的 8 条 A 类案例中,已有 8 条完成本地证据快照;0 条仍在快照队列。

SNAPSHOT 100%
已快照 8

可用 manifest 复核

待快照 0

等待落盘

部分快照 0

需补抓

需关注 0

需人工复核

证据完整度

5 / 5

模型卡 OK

Feishu Bitable

官方/厂商来源 OK

2 个入口

A 类案例 OK

8 条

公开产物 OK

已绑定

风险备注 OK

已标注

能力边界

用标签表达信号,不用标签替代证据。

查看证据方法
Coding Agent

通用/待核验

长上下文

暂无数据

研究任务

待核验

开放生态

平台/待核验

生命周期风险

厂商ByteDance Seed
厂商页ByteDance Seed
发布时间暂无数据
模型 IDseed1-5-vl
上下文暂无数据
输出暂无数据
模态文本 / 多模态待核验
推理否或官方未披露
价格官方未披露 / 暂无数据
可用平台ByteDance Seed
AA 评分暂无数据
基础模型发布时间Artificial Analysis 未披露
资料来源公开资料、厂商信息和案例库
A 类案例8

谱系位置

它在厂商路线中的位置

此页将 Seed1.5 VL 放入 ByteDance Seed 的当前 Atlas 路线中。若同厂前后代资料不足,先保留为路线追踪入口,不脑补谱系关系。

暂无数据

Seed1.5 VL 资料状态

来自厂商页/模型卡骨架 + 2 个厂商/官方/模型家族证据入口;暂无 Artificial Analysis 行。

公开档案

模型状态

Seed1.5 VL 当前标记为已有真实案例。

案例库

公开案例复核

已有 8 条可核验 A 类案例,达到完整补齐线

案例时间线

最近进入档案的 A 类案例

用采集时间呈现证据进入 Atlas 的顺序,方便复核来源新鲜度。

适合场景

  • - 作为完整模型索引和厂商路线追踪入口
  • - 通用模型对比

不适合

  • - 缺少真实案例时不要包装成推荐
  • - 缺字段时不要脑补价格、性能或上下文

真实案例

可核验 A 类案例精选

当前展示 8 条代表案例;排序优先看模型卡精选、证据可信度、快照状态和公开仓库/技术任务信号。案例总数仍为 8。

Representative Evidence

代表案例排序

当前第一代表案例是“Agent TARS uses Seed1.5-VL/Doubao Thinking Vision Pro as a Volcengine provider option for multimodal computer and browser agents”。它会优先出现在本页,是因为该案例同时进入模型卡精选、具备 100/100 的证据可信度,并带有 已快照证据。

打开第一代表案例
排序池 8 条

模型卡精选优先

证据可信度 100/100

A+ 完整链路

快照状态 已快照

archive/evidence/seed1-5-vl-agent-tars-volcengine-provider/manifest.json

技术信号 命中

Coding / repo / agent

bytedance / UI-TARS-desktop and Age… 使用 Seed1.5 VL 处理研究分析和报告生成

bytedance / UI-TARS-desktop and Agent TARS · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:28:16Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

bytedance / UI-TARS-desktop and Agent TARS 公开的研究与报告生成案例,来源为 公开代码库,复核于 2026-06-27T08:28:16Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕研究分析和报告生成的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Run Agent TARS CLI with provider volcengine and model doubao-1-5-thinking-vision-pro-250428 to execute multimodal agent tasks across terminal, computer, browser and MCP tools, including showcased booking, chart-generati…

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The public Agent TARS artifact ships CLI/Web UI documentation, showcase videos, and quick-start commands that bind Seed1.5-VL's Volcano Engine model ID to real GUI-agent and browser-operator workflows.

模型作用:Seed1.5 VL 在该案例中承担研究分析和报告生成相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL contributes multimodal vision and reasoning for GUI/browser-agent perception and planning when users choose the Volcengine provider model ID.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is official open-source product documentation and snapshots that explicitly reference doubao-1-5-thinking-vision-pro-250428; because the project is ByteDance-affiliated, it is less independent than third-party …

原始记录:Agent TARS uses Seed1.5-VL/Doubao Thinking Vision Pro as a Volcengine provider option for multimodal computer and browser agents

已有真实案例 研究与报告生成公开代码库A 类可核验real_case auto_approved 进入模型卡精选

DSeaStar / ai-diary 使用 Seed1.5 VL 处理研究分析和报告生成

DSeaStar / ai-diary · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:28:16Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

DSeaStar / ai-diary 公开的研究与报告生成案例,来源为 公开代码库,复核于 2026-06-27T08:28:16Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕研究分析和报告生成的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Automatically take hourly desktop screenshots, upload the image, ask doubao-1-5-thinking-vision-pro-250428 to summarize what appears on screen and infer the user's activity, then turn the analysis into diary content.

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The application saves generated diary text under a diaries directory and can sync the generated content and screenshot URL into Notion when configured.

模型作用:Seed1.5 VL 在该案例中承担研究分析和报告生成相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL provides multimodal screenshot understanding: it reads the uploaded screenshot and produces the concise visual activity summary that becomes the input for diary generation.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is public source code plus README describing the Doubao screenshot-analysis workflow; repository does not publish production usage metrics.

原始记录:AI Diary uses Seed1.5-VL/Doubao Thinking Vision Pro to analyze screenshots and generate diary entries

已有真实案例 研究与报告生成公开代码库A 类可核验real_case auto_approved 进入模型卡精选

mocehu / image2form_llm 使用 Seed1.5 VL 处理文档理解和结构化处理

mocehu / image2form_llm · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:28:16Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

mocehu / image2form_llm 公开的文档理解与处理案例,来源为 公开代码库,复核于 2026-06-27T08:28:16Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:Use a multimodal LLM application to convert pictures into form data; the model store exposes doubao-1-5-thinking-vision-pro-250428 under the '豆包多模态深度思考模型' group as a selectable thinking/vision model.

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The application artifact is a public Vue/Pinia project whose predefined model list lets users select Seed1.5-VL/Doubao Thinking Vision Pro for image-to-form conversion workflows.

模型作用:Seed1.5 VL 在该案例中承担文档理解和结构化处理相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL supplies the multimodal image-understanding and reasoning model option for extracting structured form content from images.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is a public application repository with an explicit model ID in the product model store and a repo description stating the image-to-form task; runtime adoption is not quantified.

原始记录:image2form_llm offers Seed1.5-VL/Doubao Thinking Vision Pro as a predefined model for converting images into forms

已有真实案例 文档理解与处理公开代码库A 类可核验real_case auto_approved 进入模型卡精选

Openmyeyesxz / Image-Renamd_By_AI-O… 使用 Seed1.5 VL 处理软件工程任务执行

Openmyeyesxz / Image-Renamd_By_AI-OCR · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:28:16Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

Openmyeyesxz / Image-Renamd_By_AI-OCR 公开的代码代理与软件工程案例,来源为 公开代码库,复核于 2026-06-27T08:28:16Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Detect rigid field-trial sample labels with YOLO, crop the label region, send the crop to Ark using model doubao-1-5-thinking-vision-pro-250428, OCR the sample name/code, and batch rename images for phenotypic data coll…

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The public repo provides a command-line workflow that writes rename_mapping.csv and moves successfully processed images into renamed output folders using OCR-derived filenames such as sample-code numbered names.

模型作用:Seed1.5 VL 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL, exposed as doubao-1-5-thinking-vision-pro-250428 on Volcano Engine Ark, performs the visual OCR step on cropped label images and returns the text used as the canonical filename base.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is a public GitHub project and code path explicitly sets DEFAULT_ARK_MODEL to doubao-1-5-thinking-vision-pro-250428; usage scale is not independently reported.

原始记录:Image-Renamd_By_AI-OCR uses Seed1.5-VL/Doubao Thinking Vision Pro for field-trial label OCR and batch image renaming

已有真实案例 代码代理与软件工程公开代码库A 类可核验real_case auto_approved 进入模型卡精选

schxar / LibrarySearch 使用 Seed1.5 VL 处理研究分析和报告生成

schxar / LibrarySearch · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:28:16Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

schxar / LibrarySearch 公开的研究与报告生成案例,来源为 公开代码库,复核于 2026-06-27T08:28:16Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕研究分析和报告生成的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Provide a Flask VLM service in a broader AI chat, book-search and multimedia-processing platform; the service accepts base64 image data and a prompt, calls Ark with model doubao-1-5-thinking-vision-pro-250428, and retur…

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The code returns structured JSON containing the image-analysis text or validation/error messages, enabling the frontend/chat service to handle image inputs alongside text and search workflows.

模型作用:Seed1.5 VL 在该案例中承担研究分析和报告生成相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL is the visual reasoning backend for uploaded images, converting image_url plus prompt inputs into natural-language analysis results.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence combines README project overview and public VLM service code that hard-codes doubao-1-5-thinking-vision-pro-250428; deployment scale is not stated.

原始记录:LibrarySearch uses Seed1.5-VL/Doubao Thinking Vision Pro for image-content analysis inside an AI chat and library-search platform

已有真实案例 研究与报告生成公开代码库A 类可核验real_case auto_approved 进入模型卡精选

langgenius / Dify 使用 Seed1.5 VL 处理软件工程任务执行

langgenius / Dify · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:23:39Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

langgenius / Dify 公开的代码代理与软件工程案例,来源为 公开代码库,复核于 2026-06-27T08:23:39Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Dify's official plugin repository added the Doubao-1.5-thinking-vision-pro model to the Volcengine MaaS provider so Dify users can select the ByteDance vision-thinking model inside application workflows.

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The merged PR is titled 'feat: add Doubao-1.5-thinking-vision-pro model' and records the model addition in Dify's official plugins artifact for provider-based model invocation.

模型作用:Seed1.5 VL 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL contributes vision-language reasoning capability to Dify apps through the Volcengine MaaS provider, enabling workflows that need image understanding plus text reasoning.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is an official plugin integration PR, not an end-customer narrative; it is accepted because it is a public, merged software artifact enabling concrete use of the model in Dify.

原始记录:Dify official Volcengine MaaS plugin adds Doubao-1.5-thinking-vision-pro for multimodal app workflows

已有真实案例 代码代理与软件工程公开代码库A 类可核验real_case auto_approved 进入模型卡精选

web-infra-dev / Midscene.js 使用 Seed1.5 VL 处理代码审查和测试生成

web-infra-dev / Midscene.js · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:23:39Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

web-infra-dev / Midscene.js 公开的代码审查与测试案例,来源为 公开代码库,复核于 2026-06-27T08:23:39Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕代码审查和测试生成的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Midscene's documentation update introduced configuration for Doubao-1.5-thinking-vision-pro from Volcano Engine in an open-source UI automation framework that writes tests in natural language and drives interfaces throu…

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The PR summary states that documentation was updated to introduce and configure Doubao-1.5-thinking-vision-pro in English and Chinese model-provider guides; the Midscene artifact is a public vision-driven UI testing and…

模型作用:Seed1.5 VL 在该案例中承担代码审查和测试生成相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL acts as the multimodal model behind Midscene-style visual UI interpretation, allowing natural-language test steps to be grounded in screenshots and page state.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is a documentation/configuration PR mirrored through a public fork rather than an upstream customer story; it is included because it binds the model to a concrete open-source UI automation product artifact.

原始记录:Midscene documents Doubao-1.5-thinking-vision-pro for vision-driven UI automation

已有真实案例 代码审查与测试公开代码库A 类可核验real_case auto_approved 进入模型卡精选

Skyvern-AI 使用 Seed1.5 VL 处理多模态内容处理

Skyvern-AI · Seed1.5 VL

A
厂商:ByteDance Seed 模型:Seed1.5 VL 来源平台:github 最后复核:2026-06-27T08:23:39Z 证据快照:已快照 · 2 / 2 个证据目标 归档说明:已完成快照:2 / 2 个证据目标,可用本地 manifest 复核证据指纹。
证据可信度 100/100

A+ 完整链路 · 代码仓库证据

原始证据1 个公开产物复核通过代码仓库证据

Skyvern-AI 公开的多模态生成与理解案例,来源为 公开代码库,复核于 2026-06-27T08:23:39Z。 Model Atlas 将它标记为 A 类证据,因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。

任务:这是一个围绕多模态内容处理的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Skyvern added a UI-TARS client path that calls the Volcano Engine endpoint with model doubao-1-5-thinking-vision-pro-250428 and uses it to generate actions from scraped browser pages for web automation tasks.

公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:The merged Skyvern PR exposes UI-TARS action generation in the Skyvern codebase; the evidence page shows the exact Seed1.5-VL-linked model id and code comments describing generation of actions using UI-TARS (Seed1.5-VL).

模型作用:Seed1.5 VL 在该案例中承担多模态内容处理相关的生成、分析、编排或实现角色。 原始资料写作:Seed1.5-VL supplies multimodal screen/page understanding and action selection for GUI/browser automation, turning visual page state into executable Skyvern actions.

A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。

风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is a GitHub integration PR rather than a customer story; it clearly binds the exact Volcano Engine model id to Seed1.5-VL and a concrete open-source automation artifact.

原始记录:Skyvern integrates Seed1.5-VL / Doubao thinking-vision for UI-TARS browser automation actions

已有真实案例 多模态生成与理解公开代码库A 类可核验real_case auto_approved 进入模型卡精选

数据缺口

只基于现有字段判断缺口;缺失项不会被猜测填充。

价格:待补公开资料 上下文:待补公开资料 输出:待补公开资料 官方/家族来源:已有 发布时间:待补公开资料 可用平台:已有 A 类案例:已有
  • - 补充方向:核验 价格 的官方/API/案例来源。
  • - 补充方向:核验 上下文 的官方/API/案例来源。
  • - 补充方向:核验 输出 的官方/API/案例来源。
  • - 补充方向:核验 发布时间 的官方/API/案例来源。

下一步验证建议

  • - 继续保留 A 类案例的原始证据、产物页和版本快照。
  • - 复核官方来源与模型 ID、发布时间、平台可用性是否仍一致。
  • - 把 benchmark、教程、发布文保留为背景资料,不提升为真实案例。