langgenius/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-07-21 03:55:25 -04:00
Code Issues 775 Packages Projects Releases 10 Wiki Activity

[BUG]VertexAI Plugin is broken in Cloud. #11952

New Issue
Closed
opened 2026-02-21 19:04:56 -05:00 by yindo · 24 comments
yindo commented 2026-02-21 19:04:56 -05:00
Owner
Copy Link
Copy Source

Originally created by @Sanuki-073 on GitHub (Mar 21, 2025).

Originally assigned to: @crazywoola on GitHub.

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

1.1.1

Cloud or Self Hosted

Cloud

Steps to reproduce

After installing the VertexAI plugin (version 0.0.8) from the marketplace and configuring Vertex AI in the model provider, an error occurred stating that the module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'.

Additionally, when installing version 0.0.7, although normal conversations work, the agent becomes unusable due to the following error: Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}"}

✔️ Expected Behavior

The system should operate without errors, allowing normal conversations and proper agent functionality without encountering issues such as missing attributes like 'GenerativeModel' or 'Schema'.

❌ Actual Behavior

version: 0.0.8:
Could not set up VertexAI.
Error message is "module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'. "
version: 0.0.7:
Could not use Gemini 2.0 flash in Agent.
Error message is "Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}"}"
version: 0.0.6:
is same as 0.0.7.

Originally created by @Sanuki-073 on GitHub (Mar 21, 2025). Originally assigned to: @crazywoola on GitHub. ### 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 1.1.1 ### Cloud or Self Hosted Cloud ### Steps to reproduce After installing the VertexAI plugin (version 0.0.8) from the marketplace and configuring Vertex AI in the model provider, an error occurred stating that the module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'. Additionally, when installing version 0.0.7, although normal conversations work, the agent becomes unusable due to the following error: Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {\"args\":{\"description\":\"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'\"},\"error_type\":\"InvokeError\",\"message\":\"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'\"}"} ### ✔️ Expected Behavior The system should operate without errors, allowing normal conversations and proper agent functionality without encountering issues such as missing attributes like 'GenerativeModel' or 'Schema'. ### ❌ Actual Behavior version: 0.0.8: Could not set up VertexAI. Error message is "module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'. " version: 0.0.7: Could not use Gemini 2.0 flash in Agent. Error message is "Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}"}" version: 0.0.6: is same as 0.0.7.
yindo added the 🐞 bugcloud labels 2026-02-21 19:04:56 -05:00
yindo closed this issue 2026-02-21 19:04:56 -05:00
yindo commented 2026-02-21 19:04:56 -05:00
Author
Owner
Copy Link
Copy Source

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

@Sanuki-073 Hello there! 👋 I'm here to help you with bug fixes and any questions you might have while waiting for a human maintainer. Let's squash those bugs together!

A similar issue was reported and closed in the past: Error when using VertexAI Gemini 1.5 Pro model for conversation. This issue was resolved by a pull request that fixed the agent's operation with Vertex AI Gemini models [1]. You might want to check if the changes from that pull request are applicable to your situation or if there have been any updates since then that could address the current problem.

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 (Mar 21, 2025): <!-- Greeting --> @Sanuki-073 Hello there! 👋 I'm here to help you with bug fixes and any questions you might have while waiting for a human maintainer. Let's squash those bugs together! <!-- Answer --> A similar issue was reported and closed in the past: [Error when using VertexAI Gemini 1.5 Pro model for conversation](https://github.com/langgenius/dify/issues/4981). This issue was resolved by a pull request that fixed the agent's operation with Vertex AI Gemini models <sup>[[1]](https://github.com/langgenius/dify/issues/4981)</sup>. You might want to check if the changes from that pull request are applicable to your situation or if there have been any updates since then that could address the current problem. <!-- 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/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/df751918-53d6-4167-9ebf-801f8714fec9?feedback_type=other)</sup>
yindo commented 2026-02-21 19:04:56 -05:00
Author
Owner
Copy Link
Copy Source

@Sanuki-073 commented on GitHub (Mar 21, 2025):

Based on Dosubot advice, I investigated the issue and found that this GitHub issue appears to be related to the cause of the problem in version 0.0.7.

Also, when forcibly upgrading to version 0.0.8 after setting up with version 0.0.7, the following error occurs:
Run failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: type object 'Part' has no attribute 'from_text'"},"error_type":"InvokeError","message":"[models] Error: type object 'Part' has no attribute 'from_text'"}

@Sanuki-073 commented on GitHub (Mar 21, 2025): Based on Dosubot advice, I investigated the issue and found that [this GitHub issue](https://github.com/langgenius/dify/issues/12545) appears to be related to the cause of the problem in version 0.0.7. Also, when forcibly upgrading to version 0.0.8 after setting up with version 0.0.7, the following error occurs: Run failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: type object 'Part' has no attribute 'from_text'"},"error_type":"InvokeError","message":"[models] Error: type object 'Part' has no attribute 'from_text'"}
yindo commented 2026-02-21 19:04:56 -05:00
Author
Owner
Copy Link
Copy Source

@Sanuki-073 commented on GitHub (Mar 21, 2025):

It seems that this change is a major mistake...
VertexAI Models: replace vertexai library into google-cloud-aiplatform

@Sanuki-073 commented on GitHub (Mar 21, 2025): It seems that this change is a major mistake... [VertexAI Models: replace vertexai library into google-cloud-aiplatform](https://github.com/langgenius/dify-official-plugins/pull/500)
yindo commented 2026-02-21 19:04:56 -05:00
Author
Owner
Copy Link
Copy Source

@crazywoola commented on GitHub (Mar 21, 2025):

I noticed that you used 0.0.6 and 0.0.7, they both have some problems right?

@crazywoola commented on GitHub (Mar 21, 2025): I noticed that you used 0.0.6 and 0.0.7, they both have some problems right?
yindo commented 2026-02-21 19:04:56 -05:00
Author
Owner
Copy Link
Copy Source

@Sanuki-073 commented on GitHub (Mar 21, 2025):

@crazywoola

I noticed that you used 0.0.6 and 0.0.7, they both have some problems right?

Yes.
If I use version 0.0.7, I got a error messages in Agent block:
Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}"}

Image
@Sanuki-073 commented on GitHub (Mar 21, 2025): @crazywoola >I noticed that you used 0.0.6 and 0.0.7, they both have some problems right? Yes. If I use version 0.0.7, I got a error messages in Agent block: Run failed: Failed to transform agent message: PluginInvokeError: {"args":{},"error_type":"Exception","message":"read llm model failed: request failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}"} <img width="1462" alt="Image" src="https://github.com/user-attachments/assets/66fe7e93-90b0-4c8a-b9f2-1c2125d9dd91" />
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@KevinKangnb commented on GitHub (Mar 24, 2025):

version 1.1.0 ~ 1.1.2 All errors reported ： module 'google.cloud.aiplatform v1 hasno attribute 'GenerativeModel

@KevinKangnb commented on GitHub (Mar 24, 2025): version 1.1.0 ~ 1.1.2 All errors reported ： module 'google.cloud.aiplatform v1 hasno attribute 'GenerativeModel
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@Sanuki-073 commented on GitHub (Mar 24, 2025):

@kangjianhua
This probably isn't an issue with Dify's version, but rather a bug in the VertexAI plugin version 0.0.8.
If it's just a simple chat (not involving usage within Tool or Agent blocks), version 0.0.7 should work.

@Sanuki-073 commented on GitHub (Mar 24, 2025): @kangjianhua This probably isn't an issue with Dify's version, but rather a bug in the VertexAI plugin version 0.0.8. If it's just a simple chat (not involving usage within Tool or Agent blocks), version 0.0.7 should work.
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@david-sabata commented on GitHub (Mar 24, 2025):

I'm having the same issue on a self-hosted instance. I cannot even configure the plugin. The version is 0.0.8

{
    "code": "invalid_param",
    "message": "module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'",
    "status": 400
}

Downgrading to 0.0.6 helped

@david-sabata commented on GitHub (Mar 24, 2025): I'm having the same issue on a self-hosted instance. I cannot even configure the plugin. The version is `0.0.8` ```json { "code": "invalid_param", "message": "module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'", "status": 400 } ``` Downgrading to `0.0.6` helped
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@abbottdev commented on GitHub (Mar 25, 2025):

also just ran into this with the following logs running on a local setup:

e9d-4bc7-44c5-8923-0367ea63debf/dispatch/llm/invoke"
2025-03-25 16:24:01 api-1            | 2025-03-25 16:24:01.475 DEBUG [Thread-1491 (_generate_worker)] [connectionpool.py:544] - http://plugin_daemon:5002 "POST /plugin/f808fe9d-4bc7-44c5-8923-0367ea63debf/dispatch/llm/invoke HTTP/1.1" 200 None
2025-03-25 16:24:01 api-1            | 2025-03-25 16:24:01.493 DEBUG [Dummy-1492] [based_generate_task_pipeline.py:45] - error: Run failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: type object 'Part' has no attribute 'from_text'"},"error_type":"InvokeError","message":"[models] Error: type object 'Part' has no attribute 'from_text'"}
2025-03-25 16:24:01 nginx-1          | 192.168.65.1 - - [25/Mar/2025:16:24:01 +0000] "POST /console/api/apps/c83ea2c1-4c1a-421f-bb07-8b5845c11169/advanced-chat/workflows/draft/run HTTP/1.1" 200 11752 "http://localhost/app/c83ea2c1-4c1a-421f-bb07-8b5845c11169/workflow" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/134.0.0.0 Safari/537.36" "-"

This was when trying to use the Google Vertex Cloud provider from the studio prompt setup.

I can configure the plugin correctly it seems - but just unable to use it.

@abbottdev commented on GitHub (Mar 25, 2025): also just ran into this with the following logs running on a local setup: ``` e9d-4bc7-44c5-8923-0367ea63debf/dispatch/llm/invoke" 2025-03-25 16:24:01 api-1 | 2025-03-25 16:24:01.475 DEBUG [Thread-1491 (_generate_worker)] [connectionpool.py:544] - http://plugin_daemon:5002 "POST /plugin/f808fe9d-4bc7-44c5-8923-0367ea63debf/dispatch/llm/invoke HTTP/1.1" 200 None 2025-03-25 16:24:01 api-1 | 2025-03-25 16:24:01.493 DEBUG [Dummy-1492] [based_generate_task_pipeline.py:45] - error: Run failed: [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: type object 'Part' has no attribute 'from_text'"},"error_type":"InvokeError","message":"[models] Error: type object 'Part' has no attribute 'from_text'"} 2025-03-25 16:24:01 nginx-1 | 192.168.65.1 - - [25/Mar/2025:16:24:01 +0000] "POST /console/api/apps/c83ea2c1-4c1a-421f-bb07-8b5845c11169/advanced-chat/workflows/draft/run HTTP/1.1" 200 11752 "http://localhost/app/c83ea2c1-4c1a-421f-bb07-8b5845c11169/workflow" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/134.0.0.0 Safari/537.36" "-" ``` This was when trying to use the Google Vertex Cloud provider from the studio prompt setup. I can configure the plugin correctly it seems - but just unable to use it.
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Mar 25, 2025):

To downgrade the VertexAI plugin to version 0.0.6, you can follow these steps:

  1. Uninstall the Current Version: First, remove the existing version of the VertexAI plugin. This can typically be done through your package manager or by manually deleting the plugin files, depending on how it was installed.

  2. Install Version 0.0.6: You can install the specific version by specifying it in your package manager. For example, if you're using pip, you can run:

    pip install vertexai==0.0.6

    Make sure to replace vertexai with the correct package name if it's different.

  3. Verify Installation: After installation, verify that the correct version is installed by checking the plugin version. You can do this by running:

    pip show vertexai

    This should display version 0.0.6.

  4. Restart Your Application: Once the correct version is installed, restart your application or service to ensure the changes take effect.

If you encounter any issues during the downgrade process, please let me know!

Additionally, could you provide more details about your current setup, such as the environment you're using (e.g., operating system, cloud provider), and any specific configurations or settings you have applied? This information might help in understanding the issue better and finding a solution.

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 (Mar 25, 2025): <!-- Answer --> To downgrade the VertexAI plugin to version `0.0.6`, you can follow these steps: 1. **Uninstall the Current Version**: First, remove the existing version of the VertexAI plugin. This can typically be done through your package manager or by manually deleting the plugin files, depending on how it was installed. 2. **Install Version 0.0.6**: You can install the specific version by specifying it in your package manager. For example, if you're using pip, you can run: ```bash pip install vertexai==0.0.6 ``` Make sure to replace `vertexai` with the correct package name if it's different. 3. **Verify Installation**: After installation, verify that the correct version is installed by checking the plugin version. You can do this by running: ```bash pip show vertexai ``` This should display version `0.0.6`. 4. **Restart Your Application**: Once the correct version is installed, restart your application or service to ensure the changes take effect. If you encounter any issues during the downgrade process, please let me know! Additionally, could you provide more details about your current setup, such as the environment you're using (e.g., operating system, cloud provider), and any specific configurations or settings you have applied? This information might help in understanding the issue better and finding a solution. <!-- 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/8b241463-32af-4266-8160-7eca9e503515?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/8b241463-32af-4266-8160-7eca9e503515?feedback_type=other)</sup>
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@xiaobinsheng001 commented on GitHub (Mar 26, 2025):

I'm having the same issue on a self-hosted instance. I cannot even configure the plugin. The version is 1.1.3
error：AttributeError: module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'

@xiaobinsheng001 commented on GitHub (Mar 26, 2025): I'm having the same issue on a self-hosted instance. I cannot even configure the plugin. The version is 1.1.3 error：AttributeError: module 'google.cloud.aiplatform_v1' has no attribute 'GenerativeModel'
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@Sanuki-073 commented on GitHub (Mar 26, 2025):

I discovered that AgentBlock's agentic strategy encounters an error when using function calling, but works without error with ReAct in version 0.0.7.

@Sanuki-073 commented on GitHub (Mar 26, 2025): I discovered that AgentBlock's agentic strategy encounters an error when using function calling, but works without error with ReAct in version 0.0.7.
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@ytjhai commented on GitHub (Apr 2, 2025):

I started seeing this after upgrading from 0.15.3 to 1.1.x, will likely have to downgrade back to 0.15.5 because of this issue. Makes the app unusable because all embeddings are done with Vertex AI.

Edit: Downloading 0.0.6 from dify marketplace, then uploading that to hosted instance resolved this issue.

@ytjhai commented on GitHub (Apr 2, 2025): I started seeing this after upgrading from 0.15.3 to 1.1.x, will likely have to downgrade back to 0.15.5 because of this issue. Makes the app unusable because all embeddings are done with Vertex AI. Edit: Downloading 0.0.6 from dify marketplace, then uploading that to hosted instance resolved this issue.
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@fdb02983rhy commented on GitHub (Apr 3, 2025):

FYI

@fdb02983rhy commented on GitHub (Apr 3, 2025): FYI - https://github.com/langgenius/dify-official-plugins/pull/651
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@david-sabata commented on GitHub (Apr 11, 2025):

Still broken for me. Self-hosted, Dify 1.2.0, plugin 0.0.9. . This is for agent; simple chat works fone.

None of the Gemini models work. Claude 3.5. Sonnect v2 works fine

[vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}
@david-sabata commented on GitHub (Apr 11, 2025): Still broken for me. Self-hosted, Dify 1.2.0, plugin 0.0.9. . **This is for agent; simple chat works fone.** None of the Gemini models work. Claude 3.5. Sonnect v2 works fine ``` [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"} ```
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@Ken-Mok commented on GitHub (Apr 14, 2025):

have the same issue, Self-hosted, Dify 1.1.3, plugin 0.0.9

Still broken for me. Self-hosted, Dify 1.2.0, plugin 0.0.9. . This is for agent; simple chat works fone.

None of the Gemini models work. Claude 3.5. Sonnect v2 works fine

[vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}
@Ken-Mok commented on GitHub (Apr 14, 2025): have the same issue, Self-hosted, Dify 1.1.3, plugin 0.0.9 > Still broken for me. Self-hosted, Dify 1.2.0, plugin 0.0.9. . **This is for agent; simple chat works fone.** > > None of the Gemini models work. Claude 3.5. Sonnect v2 works fine > > ``` > [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"} > ```
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@dpleus commented on GitHub (Apr 14, 2025):

Same here, using Cloud. Really blocking some of our projects.

@dpleus commented on GitHub (Apr 14, 2025): Same here, using Cloud. Really blocking some of our projects.
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@t-ikeda-mm commented on GitHub (May 13, 2025):

Even with the latest Vertex AI plugin (v0.0.14), Dify 1.3.1 . I’m still running into errors and the newest models don’t appear. Is anyone actually able to get it working?

@t-ikeda-mm commented on GitHub (May 13, 2025): Even with the latest Vertex AI plugin (v0.0.14), Dify 1.3.1 . I’m still running into errors and the newest models don’t appear. Is anyone actually able to get it working?
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@crazywoola commented on GitHub (May 13, 2025):

Can you paste the error messages here?

@crazywoola commented on GitHub (May 13, 2025): Can you paste the error messages here?
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@jortegac commented on GitHub (May 14, 2025):

I'm using self-hosted with 1.2.0.

VertexAI plugin 0.0.14 still cannot use tools. I get the error with any Gemini 2.0 or 2.5 variant

[vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"}
@jortegac commented on GitHub (May 14, 2025): I'm using self-hosted with 1.2.0. VertexAI plugin 0.0.14 still cannot use tools. I get the error with any Gemini 2.0 or 2.5 variant ``` [vertex_ai] Error: PluginInvokeError: {"args":{"description":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"},"error_type":"InvokeError","message":"[models] Error: module 'vertexai.generative_models' has no attribute 'Schema'"} ```
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@t-ikeda-mm commented on GitHub (May 15, 2025):

@crazywoola
I'm very satisfied with the excellent features of the community edition.
Although I'm mainly using Azure OpenAI and thus not directly affected, I would still like to be able to use Gemini if possible.

I've consulted several times about the plugin error with VertexAI, but it seems that the issue might not be with VertexAI itself.
Rather, I suspect that updates to the plugin are not being properly reflected in the configuration database.

Even after updating the VertexAI plugin to the latest version 0.0.14, the model list doesn't get updated, and models like Gemini 2.5 are not displayed.
It seems like this configuration update issue is happening only in my environment.

My setup is hosted on AWS EC2, but the PostgreDB is on AWS RDS rather than directly on EC2.
Currently, my Dify version is 1.3.1, but the original database was created back when the version was below 0.15.0 and has been updated repeatedly since then.
I suspect that this update history might be affecting the behavior.

I have already aligned the app/.env and web/.env.local files with the latest versions.

I’m considering starting completely fresh by rebuilding everything from the database up, using the latest version.
If that resolves the issue, then it’s likely not a problem with the plugin itself — and I’ll report back accordingly.
Image

Here is the error I'm encountering: if I assign any tool to an agent using Gemini and send a request, I receive an error like the one below.

Image

I sincerely appreciate this wonderful product.

reference Issue #12545
https://github.com/langgenius/dify/issues/12545

@t-ikeda-mm commented on GitHub (May 15, 2025): @crazywoola I'm very satisfied with the excellent features of the community edition. Although I'm mainly using Azure OpenAI and thus not directly affected, I would still like to be able to use Gemini if possible. I've consulted several times about the plugin error with VertexAI, but it seems that the issue might not be with VertexAI itself. Rather, I suspect that updates to the plugin are not being properly reflected in the configuration database. Even after updating the VertexAI plugin to the latest version 0.0.14, the model list doesn't get updated, and models like Gemini 2.5 are not displayed. It seems like this configuration update issue is happening only in my environment. My setup is hosted on AWS EC2, but the PostgreDB is on AWS RDS rather than directly on EC2. Currently, my Dify version is 1.3.1, but the original database was created back when the version was below 0.15.0 and has been updated repeatedly since then. I suspect that this update history might be affecting the behavior. I have already aligned the app/.env and web/.env.local files with the latest versions. I’m considering starting completely fresh by rebuilding everything from the database up, using the latest version. If that resolves the issue, then it’s likely not a problem with the plugin itself — and I’ll report back accordingly. ![Image](https://github.com/user-attachments/assets/5309a2d1-cf0b-4996-97ed-ccc9cd6d7fa4) Here is the error I'm encountering: if I assign any tool to an agent using Gemini and send a request, I receive an error like the one below. ![Image](https://github.com/user-attachments/assets/998cff01-17d7-460c-a85b-31b0cf0a11f6) I sincerely appreciate this wonderful product. reference Issue #12545 https://github.com/langgenius/dify/issues/12545
yindo commented 2026-02-21 19:04:57 -05:00
Author
Owner
Copy Link
Copy Source

@fdb02983rhy commented on GitHub (May 15, 2025):

@crazywoola
I'm very satisfied with the excellent features of the community edition.
Although I'm mainly using Azure OpenAI and thus not directly affected, I would still like to be able to use Gemini if possible.

I've consulted several times about the plugin error with VertexAI, but it seems that the issue might not be with VertexAI itself.
Rather, I suspect that updates to the plugin are not being properly reflected in the configuration database.

Even after updating the VertexAI plugin to the latest version 0.0.14, the model list doesn't get updated, and models like Gemini 2.5 are not displayed.
It seems like this configuration update issue is happening only in my environment.

My setup is hosted on AWS EC2, but the PostgreDB is on AWS RDS rather than directly on EC2.
Currently, my Dify version is 1.3.1, but the original database was created back when the version was below 0.15.0 and has been updated repeatedly since then.
I suspect that this update history might be affecting the behavior.

I have already aligned the app/.env and web/.env.local files with the latest versions.

I’m considering starting completely fresh by rebuilding everything from the database up, using the latest version.
If that resolves the issue, then it’s likely not a problem with the plugin itself — and I’ll report back accordingly.
Image

Here is the error I'm encountering: if I assign any tool to an agent using Gemini and send a request, I receive an error like the one below.

Image

I sincerely appreciate this wonderful product.

reference Issue #12545
https://github.com/langgenius/dify/issues/12545

Its a known issue. The plugin became buggy since a couples of versions ago. I was trying to reconstruct the code but got overload error frequently then I decided to stop.
https://github.com/langgenius/dify-official-plugins/pull/726

@fdb02983rhy commented on GitHub (May 15, 2025): > @crazywoola > I'm very satisfied with the excellent features of the community edition. > Although I'm mainly using Azure OpenAI and thus not directly affected, I would still like to be able to use Gemini if possible. > > I've consulted several times about the plugin error with VertexAI, but it seems that the issue might not be with VertexAI itself. > Rather, I suspect that updates to the plugin are not being properly reflected in the configuration database. > > Even after updating the VertexAI plugin to the latest version 0.0.14, the model list doesn't get updated, and models like Gemini 2.5 are not displayed. > It seems like this configuration update issue is happening only in my environment. > > My setup is hosted on AWS EC2, but the PostgreDB is on AWS RDS rather than directly on EC2. > Currently, my Dify version is 1.3.1, but the original database was created back when the version was below 0.15.0 and has been updated repeatedly since then. > I suspect that this update history might be affecting the behavior. > > I have already aligned the app/.env and web/.env.local files with the latest versions. > > I’m considering starting completely fresh by rebuilding everything from the database up, using the latest version. > If that resolves the issue, then it’s likely not a problem with the plugin itself — and I’ll report back accordingly. > ![Image](https://github.com/user-attachments/assets/5309a2d1-cf0b-4996-97ed-ccc9cd6d7fa4) > > Here is the error I'm encountering: if I assign any tool to an agent using Gemini and send a request, I receive an error like the one below. > > ![Image](https://github.com/user-attachments/assets/998cff01-17d7-460c-a85b-31b0cf0a11f6) > > I sincerely appreciate this wonderful product. > > reference Issue #12545 > https://github.com/langgenius/dify/issues/12545 Its a known issue. The plugin became buggy since a couples of versions ago. I was trying to reconstruct the code but got overload error frequently then I decided to stop. https://github.com/langgenius/dify-official-plugins/pull/726
yindo commented 2026-02-21 19:04:58 -05:00
Author
Owner
Copy Link
Copy Source

@t-ikeda-mm commented on GitHub (Jun 9, 2025):

Finally, Dify ver 1.4.1 and vertexAI plugin 0.0.18 , I find this error completely solved!
Thank you very much!

@t-ikeda-mm commented on GitHub (Jun 9, 2025): Finally, Dify ver 1.4.1 and vertexAI plugin 0.0.18 , I find this error completely solved! Thank you very much!
yindo commented 2026-02-21 19:04:58 -05:00
Author
Owner
Copy Link
Copy Source

@ytnobody commented on GitHub (Jun 17, 2025):

Today I encountered similar issue with Dify 1.2.0 and Vertex AI Plugin 0.0.19. And I downgraded Vertex AI Plugin as 0.0.16, then I got dify works well.

@ytnobody commented on GitHub (Jun 17, 2025): Today I encountered similar issue with Dify 1.2.0 and Vertex AI Plugin 0.0.19. And I downgraded Vertex AI Plugin as 0.0.16, then I got dify works well.
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 cloud 🐞 bug
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#11952
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?