langgenius/dify-plugin-sdks
1
0
Fork 0
mirror of https://github.com/langgenius/dify-plugin-sdks.git synced 2026-07-21 18:05:30 -04:00
Code Issues 58 Packages Projects Releases 10 Wiki Activity

[PR #235] [MERGED] feat: Configurable Ping Max Tokens #246

New Issue
Closed
opened 2026-02-15 21:16:24 -05:00 by yindo · 0 comments
yindo commented 2026-02-15 21:16:24 -05:00
Owner
Copy Link
Copy Source

📋 Pull Request Information

Original PR: https://github.com/langgenius/dify-plugin-sdks/pull/235
Author: @Bytechoreographer
Created: 11/4/2025
Status: ✅ Merged
Merged: 12/8/2025
Merged by: @Mairuis

Base: main ← Head: feat/increase_ping_max_tokens

📝 Commits (1)

  • a5c44ce fix: increase max_tokens from 5 to 16 for some specific models like gpt-5-codex in ping test, add PING_MAX_TOKENS configuration for model validation ping requests

📊 Changes

2 files changed (+12 additions, -2 deletions)

View changed files

📝 python/dify_plugin/config/config.py (+5 -0)
📝 python/dify_plugin/interfaces/model/openai_compatible/llm.py (+7 -2)

📄 Description

Feature: Configurable Ping Max Tokens

Overview

Changed the max_tokens parameter in OpenAI-compatible model validation (ping) requests from a hard-coded value to an environment variable-based configuration, improving flexibility and meeting different provider requirements.

Background

Some OpenAI-compatible model providers—such as OpenRouter (when wrapping other vendors or offering models like gpt-5-codex)—either require the max_tokens parameter in validation requests to be at least 16 (resulting in validation failure if not met) or enforce a minimum output token limit overall (returning a 400 error for very small values like 5). However, the previous implementation hard-coded this max_tokens value as 5, which lacked flexibility.
image

Fixes #225

Changes

1. Configuration File Changes

File: python/dify_plugin/config/config.py

Added a new environment variable configuration in the DifyPluginEnv class:

PING_MAX_TOKENS: int = Field(
    default=16,
    description="Maximum tokens for model validation ping request. Some providers require at least 16 tokens.",
)

Notes:

  • Default value set to 16, meeting the minimum requirement for most providers
  • Can be customized via the .env file
  • Clear naming with detailed description

2. Implementation Code Changes

File: python/dify_plugin/interfaces/model/openai_compatible/llm.py

2.1 Import Configuration Class

Added at the beginning of the file:

from dify_plugin.config.config import DifyPluginEnv

2.2 Modified validate_credentials Method

In the validate_credentials method:

  1. Added configuration reading logic:
# Load ping max tokens from environment variable
config = DifyPluginEnv()
ping_max_tokens = config.PING_MAX_TOKENS
  1. Replaced hard-coded max_tokens value:
# Before
data = {"model": credentials.get("endpoint_model_name", model), "max_tokens": 5}

# After
data = {"model": credentials.get("endpoint_model_name", model), "max_tokens": ping_max_tokens}
  1. Also fixed the hard-coded value in stream mode:
# Before
if stream_mode_auth == "use":
    data["stream"] = True
    data["max_tokens"] = 10

# After
if stream_mode_auth == "use":
    data["stream"] = True
    data["max_tokens"] = ping_max_tokens

Usage

1. Using Default Value (Recommended)

No configuration needed. The system will use the default value of 16.

2. Custom Configuration

Add to the .env file in the project root directory:

PING_MAX_TOKENS=32

Or set other values based on actual needs, for example:

# Some special providers may require larger values
PING_MAX_TOKENS=64

Compatibility Notes

  • Backward Compatible: Changes are fully backward compatible and do not affect existing functionality
  • Default Behavior Change: Default value changed from 5 to 16, meeting minimum requirements for most providers
  • Flexible Configuration: Supports customization via environment variables for different scenarios

Related Files

  • python/dify_plugin/config/config.py - Configuration definition
  • python/dify_plugin/interfaces/model/openai_compatible/llm.py - Implementation code

Modification Date

2025-01-06

Pull Request Checklist

Thank you for your contribution! Before submitting your PR, please make sure you have completed the following checks:

Compatibility Check

  • [✔] I have checked whether this change affects the backward compatibility of the plugin declared in README.md
  • [✔] I have checked whether this change affects the forward compatibility of the plugin declared in README.md
  • [✔] If this change introduces a breaking change, I have discussed it with the project maintainer and specified the release version in the README.md
  • [✔] I have described the compatibility impact and the corresponding version number in the PR description
  • [✔] I have checked whether the plugin version is updated in the README.md

Available Checks

  • [✔] Code has passed local tests
  • [✔] Relevant documentation has been updated (if necessary)

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/langgenius/dify-plugin-sdks/pull/235 **Author:** [@Bytechoreographer](https://github.com/Bytechoreographer) **Created:** 11/4/2025 **Status:** ✅ Merged **Merged:** 12/8/2025 **Merged by:** [@Mairuis](https://github.com/Mairuis) **Base:** `main` ← **Head:** `feat/increase_ping_max_tokens` --- ### 📝 Commits (1) - [`a5c44ce`](https://github.com/langgenius/dify-plugin-sdks/commit/a5c44ce5a4240c5cf43c8cf3f50553907de53b77) fix: increase max_tokens from 5 to 16 for some specific models like gpt-5-codex in ping test, add PING_MAX_TOKENS configuration for model validation ping requests ### 📊 Changes **2 files changed** (+12 additions, -2 deletions) <details> <summary>View changed files</summary> 📝 `python/dify_plugin/config/config.py` (+5 -0) 📝 `python/dify_plugin/interfaces/model/openai_compatible/llm.py` (+7 -2) </details> ### 📄 Description # Feature: Configurable Ping Max Tokens ## Overview Changed the `max_tokens` parameter in OpenAI-compatible model validation (ping) requests from a hard-coded value to an environment variable-based configuration, improving flexibility and meeting different provider requirements. ## Background Some OpenAI-compatible model providers—such as OpenRouter (when wrapping other vendors or offering models like gpt-5-codex)—either require the max_tokens parameter in validation requests to be at least 16 (resulting in validation failure if not met) or enforce a minimum output token limit overall (returning a 400 error for very small values like 5). However, the previous implementation hard-coded this max_tokens value as 5, which lacked flexibility. <img width="1014" height="955" alt="image" src="https://github.com/user-attachments/assets/1ecb10ce-1182-49b4-b38f-507652a70ce2" /> Fixes #225 ## Changes ### 1. Configuration File Changes **File:** `python/dify_plugin/config/config.py` Added a new environment variable configuration in the `DifyPluginEnv` class: ```python PING_MAX_TOKENS: int = Field( default=16, description="Maximum tokens for model validation ping request. Some providers require at least 16 tokens.", ) ``` **Notes:** - Default value set to 16, meeting the minimum requirement for most providers - Can be customized via the `.env` file - Clear naming with detailed description ### 2. Implementation Code Changes **File:** `python/dify_plugin/interfaces/model/openai_compatible/llm.py` #### 2.1 Import Configuration Class Added at the beginning of the file: ```python from dify_plugin.config.config import DifyPluginEnv ``` #### 2.2 Modified validate_credentials Method In the `validate_credentials` method: 1. Added configuration reading logic: ```python # Load ping max tokens from environment variable config = DifyPluginEnv() ping_max_tokens = config.PING_MAX_TOKENS ``` 2. Replaced hard-coded max_tokens value: ```python # Before data = {"model": credentials.get("endpoint_model_name", model), "max_tokens": 5} # After data = {"model": credentials.get("endpoint_model_name", model), "max_tokens": ping_max_tokens} ``` 3. Also fixed the hard-coded value in stream mode: ```python # Before if stream_mode_auth == "use": data["stream"] = True data["max_tokens"] = 10 # After if stream_mode_auth == "use": data["stream"] = True data["max_tokens"] = ping_max_tokens ``` ## Usage ### 1. Using Default Value (Recommended) No configuration needed. The system will use the default value of 16. ### 2. Custom Configuration Add to the `.env` file in the project root directory: ```bash PING_MAX_TOKENS=32 ``` Or set other values based on actual needs, for example: ```bash # Some special providers may require larger values PING_MAX_TOKENS=64 ``` ## Compatibility Notes - **Backward Compatible**: Changes are fully backward compatible and do not affect existing functionality - **Default Behavior Change**: Default value changed from 5 to 16, meeting minimum requirements for most providers - **Flexible Configuration**: Supports customization via environment variables for different scenarios ## Related Files - `python/dify_plugin/config/config.py` - Configuration definition - `python/dify_plugin/interfaces/model/openai_compatible/llm.py` - Implementation code ## Modification Date 2025-01-06 # Pull Request Checklist Thank you for your contribution! Before submitting your PR, please make sure you have completed the following checks: ## Compatibility Check - [✔] I have checked whether this change affects the **backward compatibility** of the plugin declared in `README.md` - [✔] I have checked whether this change affects the **forward compatibility** of the plugin declared in `README.md` - [✔] If this change introduces a breaking change, I have discussed it with the project maintainer and specified the release version in the `README.md` - [✔] I have described the compatibility impact and the corresponding version number in the PR description - [✔] I have checked whether the plugin version is updated in the `README.md` ## Available Checks - [✔] Code has passed local tests - [✔] Relevant documentation has been updated (if necessary) --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-02-15 21:16:24 -05:00
yindo closed this issue 2026-02-15 21:16:24 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
documentation
enhancement
pull-request
Mirrored from GitHub Pull Request
 question
No Label pull-request
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-plugin-sdks#246
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?