fix(openai): GPT-5 series models contain incorrect parameters by fdb02983rhy · Pull Request #1444 · langgenius/dify-official-plugins

fdb02983rhy · 2025-08-07T18:50:27Z

Related Issues or Context

Area.mp4

This PR contains Changes to Non-Plugin

Documentation
Other

This PR contains Changes to Non-LLM Models Plugin

I have Run Comprehensive Tests Relevant to My Changes

This PR contains Changes to LLM Models Plugin

My Changes Affect Message Flow Handling (System Messages and User→Assistant Turn-Taking)

My Changes Affect Tool Interaction Flow (Multi-Round Usage and Output Handling, for both Agent App and Agent Node)

My Changes Affect Multimodal Input Handling (Images, PDFs, Audio, Video, etc.)

My Changes Affect Multimodal Output Generation (Images, Audio, Video, etc.)

My Changes Affect Structured Output Format (JSON, XML, etc.)

My Changes Affect Token Consumption Metrics

My Changes Affect Other LLM Functionalities (Reasoning Process, Grounding, Prompt Caching, etc.)

Other Changes (Add New Models, Fix Model Parameters etc.)

Version Control (Any Changes to the Plugin Will Require Bumping the Version)

I have Bumped Up the Version in Manifest.yaml (Top-Level Version Field, Not in Meta Section)

Dify Plugin SDK Version

I have Ensured dify_plugin>=0.3.0,<0.5.0 is in requirements.txt (SDK docs)

Environment Verification (If Any Code Changes)

Local Deployment Environment

Dify Version is: , I have Tested My Changes on Local Deployment Dify with a Clean Environment That Matches the Production Configuration.

SaaS Environment

I have Tested My Changes on cloud.dify.ai with a Clean Environment That Matches the Production Configuration

- Add claude-opus-4-1-20250805.yaml with advanced features - Support for thinking mode with configurable budget - Enhanced prompt caching capabilities - Vision, tool calling, and document processing support - Bump version to 0.1.2 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>

This reverts commit 63c98db.

- Add gpt-5, gpt-5-mini, gpt-5-nano model variants - Include dated versions (2025-08-07) and latest versions - Configure model parameters and capabilities for new GPT-5 series

- Add gpt-5, gpt-5-mini, gpt-5-nano model variants to _position.yaml - Include dated versions (2025-08-07) and latest versions - Position models logically between GPT-4.1 and GPT-4o series

laipz8200 · 2025-08-07T19:17:34Z

Great! I think you can copy the parameters from o3.

laipz8200 · 2025-08-07T19:20:01Z

And...just a suggestion - would you be open to rearranging the order of the models in the position so they go from newest to oldest?

…uency_penalty (except gpt-5-chat-latest); runtime: treat o* and gpt-5* as thinking-series, map max_tokens to max_completion_tokens

Eliminated the 'json_object' option from the 'response_format' parameter in all GPT-5, GPT-5-mini, and GPT-5-nano YAML configuration files to align with updated model capabilities or requirements.

fdb02983rhy · 2025-08-07T19:52:11Z

Ok fully tested

fdb02983rhy · 2025-08-07T19:52:52Z

And...just a suggestion - would you be open to rearranging the order of the models in the position so they go from newest to oldest?

Sure thing, done.

Moved GPT-4 and related model entries to a new position in the list for improved organization and consistency.

laipz8200 · 2025-08-07T19:59:44Z

Awesome work! Thank you for your contribution!

* Add Claude Opus 4.1 model support - Add claude-opus-4-1-20250805.yaml with advanced features - Support for thinking mode with configurable budget - Enhanced prompt caching capabilities - Vision, tool calling, and document processing support - Bump version to 0.1.2 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com> * Revert "Add Claude Opus 4.1 model support" This reverts commit 63c98db. * chore: add gpt-5 series * chore: update manifest and requirements for GPT-5 series * Add GPT-5 series model configurations - Add gpt-5, gpt-5-mini, gpt-5-nano model variants - Include dated versions (2025-08-07) and latest versions - Configure model parameters and capabilities for new GPT-5 series * Add GPT-5 series models to position configuration - Add gpt-5, gpt-5-mini, gpt-5-nano model variants to _position.yaml - Include dated versions (2025-08-07) and latest versions - Position models logically between GPT-4.1 and GPT-4o series * OpenAI: GPT-5 series YAML cleanup: remove top_p/presence_penalty/frequency_penalty (except gpt-5-chat-latest); runtime: treat o* and gpt-5* as thinking-series, map max_tokens to max_completion_tokens * OpenAI: update LLM _position.yaml for GPT-5 series positioning * Update _position.yaml * Remove json_object option from GPT-5 model configs Eliminated the 'json_object' option from the 'response_format' parameter in all GPT-5, GPT-5-mini, and GPT-5-nano YAML configuration files to align with updated model capabilities or requirements. * Reorder GPT-4 model entries in _position.yaml Moved GPT-4 and related model entries to a new position in the list for improved organization and consistency. --------- Co-authored-by: Claude <noreply@anthropic.com>

fdb02983rhy and others added 5 commits August 6, 2025 22:26

Revert "Add Claude Opus 4.1 model support"

d9618fe

This reverts commit 63c98db.

Merge branch 'langgenius:main' into main

a87dd3e

chore: add gpt-5 series

6b569c7

chore: update manifest and requirements for GPT-5 series

dbb5d67

fdb02983rhy marked this pull request as ready for review August 7, 2025 18:50

fdb02983rhy added 2 commits August 8, 2025 03:54

Add GPT-5 series model configurations

595ec0a

- Add gpt-5, gpt-5-mini, gpt-5-nano model variants - Include dated versions (2025-08-07) and latest versions - Configure model parameters and capabilities for new GPT-5 series

Add GPT-5 series models to position configuration

123fb94

- Add gpt-5, gpt-5-mini, gpt-5-nano model variants to _position.yaml - Include dated versions (2025-08-07) and latest versions - Position models logically between GPT-4.1 and GPT-4o series

fdb02983rhy changed the title ~~Chore/add gpt 5 series~~ Fix/correct gpt 5 series parameters Aug 7, 2025

fdb02983rhy added 2 commits August 8, 2025 04:06

Merge branch 'langgenius:main' into main

e650a7b

Merge remote-tracking branch 'origin/main' into chore/add-gpt-5-series

42b0e04

fdb02983rhy marked this pull request as draft August 7, 2025 19:13

laipz8200 changed the title ~~Fix/correct gpt 5 series parameters~~ fix(openai): GPT-5 series models contain incorrect parameters Aug 7, 2025

nezhar reviewed Aug 7, 2025

View reviewed changes

Comment thread models/openai/models/llm/_position.yaml Outdated

fdb02983rhy added 4 commits August 8, 2025 04:34

OpenAI: GPT-5 series YAML cleanup: remove top_p/presence_penalty/freq…

d8f8915

…uency_penalty (except gpt-5-chat-latest); runtime: treat o* and gpt-5* as thinking-series, map max_tokens to max_completion_tokens

OpenAI: update LLM _position.yaml for GPT-5 series positioning

94ec694

Update _position.yaml

c3e8e70

Remove json_object option from GPT-5 model configs

7250e0c

Eliminated the 'json_object' option from the 'response_format' parameter in all GPT-5, GPT-5-mini, and GPT-5-nano YAML configuration files to align with updated model capabilities or requirements.

nezhar approved these changes Aug 7, 2025

View reviewed changes

Reorder GPT-4 model entries in _position.yaml

1a6ec7b

Moved GPT-4 and related model entries to a new position in the list for improved organization and consistency.

fdb02983rhy marked this pull request as ready for review August 7, 2025 19:54

laipz8200 approved these changes Aug 7, 2025

View reviewed changes

laipz8200 merged commit 67f5855 into langgenius:main Aug 7, 2025
2 checks passed

dosubot Bot mentioned this pull request Aug 21, 2025

Model Misinterprets Input as Empty Array in Dify with Azure OpenAI GPT-5 #1544

Closed

5 tasks

This was referenced Sep 5, 2025

Microsoft encountered errors when adding GPT-5 Nano and Mini models. #1660

Closed

There is an error in adding the GPT5mini and nano models. langgenius/dify#25333

Closed

dosubot Bot mentioned this pull request Sep 18, 2025

OpenAI plugin get_num_tokens_from_messages() is not presently implemented for model gpt-5 #1734

Closed

5 tasks

Navigation Menu

Uh oh!

fix(openai): GPT-5 series models contain incorrect parameters - #1444

fix(openai): GPT-5 series models contain incorrect parameters#1444
laipz8200 merged 14 commits into
langgenius:mainfrom
fdb02983rhy:chore/add-gpt-5-series

fdb02983rhy commented Aug 7, 2025 •

edited

Loading

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

Uh oh!

fdb02983rhy commented Aug 7, 2025

Uh oh!

fdb02983rhy commented Aug 7, 2025

Uh oh!

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Conversation

fdb02983rhy commented Aug 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Related Issues or Context

This PR contains Changes to Non-Plugin

This PR contains Changes to Non-LLM Models Plugin

This PR contains Changes to LLM Models Plugin

Version Control (Any Changes to the Plugin Will Require Bumping the Version)

Dify Plugin SDK Version

Environment Verification (If Any Code Changes)

Local Deployment Environment

SaaS Environment

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

Uh oh!

fdb02983rhy commented Aug 7, 2025

Uh oh!

fdb02983rhy commented Aug 7, 2025

Uh oh!

Uh oh!

laipz8200 commented Aug 7, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

fdb02983rhy commented Aug 7, 2025 •

edited

Loading