langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-20 09:10:05 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

Why is there a significant difference between the conversation using Dify + Ollama and the conversation directly through Ollama, even though the same model is used? #8504

New Issue
Closed
opened 2026-02-21 18:26:03 -05:00 by yindo · 2 comments
yindo commented 2026-02-21 18:26:03 -05:00
Owner
Copy Link
Copy Source

Originally created by @lizhichao999 on GitHub (Feb 21, 2025).

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.
  • I have searched for existing issues search for existing issues, including closed ones.
  • I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
  • [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
  • Please do not modify this template :) and fill in all the required fields.

Dify version

0.15.3

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Why is there a significant difference between the conversation using Dify + Ollama and the conversation directly through Ollama, even though the same model is used?

✔️ Expected Behavior

Use 2 conversations for the same result

❌ Actual Behavior

顺丰科技有限公司介绍：ollama的返回：
C:\Users\HKZC>ollama run deepseek-r1:14b

顺丰科技有限公司介绍

嗯，用户想让我介绍一下顺丰科技有限公司。首先，我需要确定他们对什么信息感兴趣。可能他们是潜在的客户、合作伙伴，或者是
投资者。

顺丰科技是顺丰控股旗下的子公司，这点很重要，应该先说明他们的背景和实力。然后，公司的业务范围包括供应链解决方案、智慧
物流、大数据分析和物联网应用。这些方面能够展示他们的技术和服务能力。

接下来，我应该提到他们的技术能力，比如无人机和无人车，这显示了他们在创新方面的投入。还要强调客户为中心的服务理念，说
明他们如何满足不同行业的需求。

最后，可以简要介绍一下顺丰科技的战略目标，即成为全球领先的智慧供应链生态服务商，这样能突出公司的长期愿景和市场定位。

顺丰科技有限公司（简称“顺丰科技”）是顺丰控股股份有限公司（以下简称“顺丰控股”或“顺丰”）的全资子公司。顺丰控股是
中国领先的快递物流综合服务提供商，而顺丰科技则是其专注于科技创新与数字化转型的核心平台。

公司概况：

  1. 母公司背景：顺丰控股成立于1993年，总部位于广东省深圳市。公司于2017年在深圳证券交易所上市（股票代码：
    002926.SZ）。顺丰控股以快递业务起家，现已发展成为涵盖快递、物流、供应链管理、跨境运输、仓储、冷链物流等多领域的综合
    物流服务商。

  2. 使命与愿景：

    • 顺丰科技致力于通过科技创新提升物流效率、降低运营成本，并推动行业向智能化和数字化方向发展。
    • 公司愿景是成为全球领先的智慧供应链生态服务商，为客户创造更大价值，为行业赋能。

核心业务领域：

  1. 供应链解决方案：顺丰科技利用大数据、人工智能和物联网等技术，为企业客户提供定制化的供应链管理服务，包括仓储优
    ���、运输规划、订单管理等。

  2. 智慧物流：通过自主研发的智能调度系统、路径优化算法以及末端配送自动化设备（如无人机、无人车），提升整个物流网
    络的效率和智能化水平。

  3. 大数据与人工智能：顺丰科技在数据采集、处理、分析及应用方面具有强大的技术能力，帮助客户实现精准预测、智能决策
    等。

  4. 物联网与区块链技术：通过IoT设备和区块链技术，实现货物追踪、溯源以及信息共享，提升供应链透明度和安全性。

  5. 跨境物流与全球化布局：顺丰科技积极推动国际化战略，在全球范围内建立物流网络，提供跨境运输、海外仓储等服务，助
    力中国企业拓展国际市场。

技术实力：

  • 顺丰科技拥有一支由数据科学家、算法工程师、系统架构师组成的技术团队，并在人工智能、大数据分析、云计算等领域拥有丰富
    的经验。
  • 公司自主研发的智能调度系统“丰智”可以实时监控和优化运输路径，提高车辆利用率和服务效率。

发展与合作：

顺丰科技不仅服务于母公司顺丰控股，还与多家外部企业建立战略合作关系，共同探索物流科技创新应用。公司积极参与行业标准制
定，并与高校、研究机构保持紧密合作，推动物流行业的技术进步。

未来展望：

顺丰科技将继续加大研发投入，拓展新兴技术在物流领域的应用场景，同时深化与合作伙伴的战略协作，共同构建智慧物流生态体系
。通过持续创新和数字化转型，顺丰科技旨在为客户提供更高效、更智能的综合物流解决方案。

如需了解更多关于顺丰科技的具体信息，可以访问公司官网或查阅其公开发布的新闻稿和财报资料。

ollama+dify的回复：
Thinking...
Alright, so I need to figure out how to introduce 顺丰科技有限公司. Hmm, where do I start? First off, I know that "顺丰" is a well-known logistics company in China, right? They handle a lot of deliveries and are pretty efficient. So, 顺丰科技有限公司 must be related to that.
Maybe it's a tech subsidiary of the main顺丰group. That makes sense because many big companies have separate tech divisions to focus on innovation and R&D. But I'm not entirely sure about that structure. Should I look into whether 顺丰科技 is indeed part of顺丰集团 or if it's a different entity altogether?

Okay, let's assume it's part of the group for now. If that's the case, what specific areas does 顺丰科技 focus on? Probably technology-related services within logistics. Could be things like supply chain management, warehouse automation, or even drone deliveries. I've heard about companies using drones for last-mile delivery in remote areas.

Also, with the rise of e-commerce in China, logistics tech is a huge area. Maybe 顺丰科技 works on optimizing delivery routes using algorithms or data analytics. That would help reduce costs and improve speed.

Another thought: they might develop their own software platforms for managing shipments, tracking packages, and integrating with other systems. Having a robust tech backbone could give them a competitive edge in the industry.

I wonder if they also work on things like smart logistics parks or using IoT devices to monitor shipments in real-time. Those technologies would definitely enhance efficiency and transparency.

But I'm not sure about all this for certain. Maybe I should check some sources or official websites to get accurate information. However, since I can't access external links right now, I'll have to work with what I know.

In summary, 顺丰科技有限公司 likely specializes in tech solutions within logistics, focusing on areas like automation, data analytics, and innovative delivery methods. They probably aim to streamline operations and provide better services to their clients.

顺丰科技有限公司是中国顺丰集团的子公司，专注于为物流行业提供高科技解决方案。公司致力于通过技术创新提升物流效率和客户体验，涵盖智能仓储、自动化配送系统以及无人机物流等领域。顺丰科技利用大数据分析和人工智能技术优化供应链管理，并开发了多个软件平台用于 shipment management 和实时 tracking。其目标是通过科技手段持续改进物流服务，保持行业领先地位。

Originally created by @lizhichao999 on GitHub (Feb 21, 2025). ### Self Checks - [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [x] [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:） - [x] Please do not modify this template :) and fill in all the required fields. ### Dify version 0.15.3 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce Why is there a significant difference between the conversation using Dify + Ollama and the conversation directly through Ollama, even though the same model is used? ### ✔️ Expected Behavior Use 2 conversations for the same result ### ❌ Actual Behavior 顺丰科技有限公司介绍：ollama的返回： C:\Users\HKZC>ollama run deepseek-r1:14b >>> 顺丰科技有限公司介绍 <think> 嗯，用户想让我介绍一下顺丰科技有限公司。首先，我需要确定他们对什么信息感兴趣。可能他们是潜在的客户、合作伙伴，或者是 投资者。 顺丰科技是顺丰控股旗下的子公司，这点很重要，应该先说明他们的背景和实力。然后，公司的业务范围包括供应链解决方案、智慧 物流、大数据分析和物联网应用。这些方面能够展示他们的技术和服务能力。 接下来，我应该提到他们的技术能力，比如无人机和无人车，这显示了他们在创新方面的投入。还要强调客户为中心的服务理念，说 明他们如何满足不同行业的需求。 最后，可以简要介绍一下顺丰科技的战略目标，即成为全球领先的智慧供应链生态服务商，这样能突出公司的长期愿景和市场定位。 </think> 顺丰科技有限公司（简称“顺丰科技”）是顺丰控股股份有限公司（以下简称“顺丰控股”或“顺丰”）的全资子公司。顺丰控股是 中国领先的快递物流综合服务提供商，而顺丰科技则是其专注于科技创新与数字化转型的核心平台。 ### 公司概况： 1. **母公司背景**：顺丰控股成立于1993年，总部位于广东省深圳市。公司于2017年在深圳证券交易所上市（股票代码： 002926.SZ）。顺丰控股以快递业务起家，现已发展成为涵盖快递、物流、供应链管理、跨境运输、仓储、冷链物流等多领域的综合 物流服务商。 2. **使命与愿景**： - 顺丰科技致力于通过科技创新提升物流效率、降低运营成本，并推动行业向智能化和数字化方向发展。 - 公司愿景是成为全球领先的智慧供应链生态服务商，为客户创造更大价值，为行业赋能。 ### 核心业务领域： 1. **供应链解决方案**：顺丰科技利用大数据、人工智能和物联网等技术，为企业客户提供定制化的供应链管理服务，包括仓储优 ���、运输规划、订单管理等。 2. **智慧物流**：通过自主研发的智能调度系统、路径优化算法以及末端配送自动化设备（如无人机、无人车），提升整个物流网 络的效率和智能化水平。 3. **大数据与人工智能**：顺丰科技在数据采集、处理、分析及应用方面具有强大的技术能力，帮助客户实现精准预测、智能决策 等。 4. **物联网与区块链技术**：通过IoT设备和区块链技术，实现货物追踪、溯源以及信息共享，提升供应链透明度和安全性。 5. **跨境物流与全球化布局**：顺丰科技积极推动国际化战略，在全球范围内建立物流网络，提供跨境运输、海外仓储等服务，助 力中国企业拓展国际市场。 ### 技术实力： - 顺丰科技拥有一支由数据科学家、算法工程师、系统架构师组成的技术团队，并在人工智能、大数据分析、云计算等领域拥有丰富 的经验。 - 公司自主研发的智能调度系统“丰智”可以实时监控和优化运输路径，提高车辆利用率和服务效率。 ### 发展与合作： 顺丰科技不仅服务于母公司顺丰控股，还与多家外部企业建立战略合作关系，共同探索物流科技创新应用。公司积极参与行业标准制 定，并与高校、研究机构保持紧密合作，推动物流行业的技术进步。 ### 未来展望： 顺丰科技将继续加大研发投入，拓展新兴技术在物流领域的应用场景，同时深化与合作伙伴的战略协作，共同构建智慧物流生态体系 。通过持续创新和数字化转型，顺丰科技旨在为客户提供更高效、更智能的综合物流解决方案。 如需了解更多关于顺丰科技的具体信息，可以访问公司官网或查阅其公开发布的新闻稿和财报资料。 ollama+dify的回复： Thinking... Alright, so I need to figure out how to introduce 顺丰科技有限公司. Hmm, where do I start? First off, I know that "顺丰" is a well-known logistics company in China, right? They handle a lot of deliveries and are pretty efficient. So, 顺丰科技有限公司 must be related to that. Maybe it's a tech subsidiary of the main顺丰group. That makes sense because many big companies have separate tech divisions to focus on innovation and R&D. But I'm not entirely sure about that structure. Should I look into whether 顺丰科技 is indeed part of顺丰集团 or if it's a different entity altogether? Okay, let's assume it's part of the group for now. If that's the case, what specific areas does 顺丰科技 focus on? Probably technology-related services within logistics. Could be things like supply chain management, warehouse automation, or even drone deliveries. I've heard about companies using drones for last-mile delivery in remote areas. Also, with the rise of e-commerce in China, logistics tech is a huge area. Maybe 顺丰科技 works on optimizing delivery routes using algorithms or data analytics. That would help reduce costs and improve speed. Another thought: they might develop their own software platforms for managing shipments, tracking packages, and integrating with other systems. Having a robust tech backbone could give them a competitive edge in the industry. I wonder if they also work on things like smart logistics parks or using IoT devices to monitor shipments in real-time. Those technologies would definitely enhance efficiency and transparency. But I'm not sure about all this for certain. Maybe I should check some sources or official websites to get accurate information. However, since I can't access external links right now, I'll have to work with what I know. In summary, 顺丰科技有限公司 likely specializes in tech solutions within logistics, focusing on areas like automation, data analytics, and innovative delivery methods. They probably aim to streamline operations and provide better services to their clients. 顺丰科技有限公司是中国顺丰集团的子公司，专注于为物流行业提供高科技解决方案。公司致力于通过技术创新提升物流效率和客户体验，涵盖智能仓储、自动化配送系统以及无人机物流等领域。顺丰科技利用大数据分析和人工智能技术优化供应链管理，并开发了多个软件平台用于 shipment management 和实时 tracking。其目标是通过科技手段持续改进物流服务，保持行业领先地位。
yindo added the 🙋‍♂️ question label 2026-02-21 18:26:03 -05:00
yindo closed this issue 2026-02-21 18:26:03 -05:00
yindo commented 2026-02-21 18:26:03 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Feb 21, 2025):

The significant difference in conversation results between using Dify with Ollama and using Ollama directly, despite using the same model, could be due to several factors related to how Dify integrates and processes model interactions. Dify might apply specific configurations or middleware that affect the model's behavior, such as context management, input/output processing, or additional layers of logic that are not present when using Ollama standalone. Unfortunately, specific details on these differences are not available in the documentation I have access to. However, known issues like connection errors when using Docker are documented, which might indirectly affect performance or output [1].

To continue talking to Dosu, mention @dosu.

Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other

@dosubot[bot] commented on GitHub (Feb 21, 2025): <!-- Answer --> The significant difference in conversation results between using Dify with Ollama and using Ollama directly, despite using the same model, could be due to several factors related to how Dify integrates and processes model interactions. Dify might apply specific configurations or middleware that affect the model's behavior, such as context management, input/output processing, or additional layers of logic that are not present when using Ollama standalone. Unfortunately, specific details on these differences are not available in the documentation I have access to. However, known issues like connection errors when using Docker are documented, which might indirectly affect performance or output <sup>[[1]](https://github.com/langgenius/dify-docs/blob/main/en/development/models-integration/ollama.md)</sup>. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.* --- <sup>Help Dosu learn! Give it feedback: [Great Response](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=other)</sup>
yindo commented 2026-02-21 18:26:03 -05:00
Author
Owner
Copy Link
Copy Source

@lizhichao999 commented on GitHub (Feb 21, 2025):

The significant difference in conversation results between using Dify with Ollama and using Ollama directly, despite using the same model, could be due to several factors related to how Dify integrates and processes model interactions. Dify might apply specific configurations or middleware that affect the model's behavior, such as context management, input/output processing, or additional layers of logic that are not present when using Ollama standalone. Unfortunately, specific details on these differences are not available in the documentation I have access to. However, known issues like connection errors when using Docker are documented, which might indirectly affect performance or output [1].

To continue talking to Dosu, mention @dosu.

Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other

I am using DeepSeek-r1:14b

@lizhichao999 commented on GitHub (Feb 21, 2025): > The significant difference in conversation results between using Dify with Ollama and using Ollama directly, despite using the same model, could be due to several factors related to how Dify integrates and processes model interactions. Dify might apply specific configurations or middleware that affect the model's behavior, such as context management, input/output processing, or additional layers of logic that are not present when using Ollama standalone. Unfortunately, specific details on these differences are not available in the documentation I have access to. However, known issues like connection errors when using Docker are documented, which might indirectly affect performance or output [[1]](https://github.com/langgenius/dify-docs/blob/main/en/development/models-integration/ollama.md). > > _To continue talking to [Dosu](https://dosu.dev), mention [@dosu](https://github.com/dosu)._ > > Help Dosu learn! Give it feedback: [Great Response](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/a17a4742-1fb9-4696-ade9-22a924f10ce4?feedback_type=other) I am using DeepSeek-r1:14b
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Knowledge Pipeline
Proposal
WIP
cloud
dependencies
feat:webapp
good first issue
outdated
performance
pull-request
Mirrored from GitHub Pull Request
 python:uv
refactor
size:L
stale
status: accepting prs
status: blocked
status: in discussion
stt & tts
vulnerablity
web
wont-fix
🌊 feat:workflow
🌚 invalid
🐞 bug
👻 feat:rag
💪 enhancement
📚 documentation
🔨 feat:tools
🙋‍♂️ question
🙏 help wanted
🤔 cant-reproduce
🤖 feat:agent
No Label 🙋‍♂️ question
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#8504
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?