[GH-ISSUE #187] [Enhancement]: vxcontrol/graphiti supports OpenAIGenericClient #74

Open
opened 2026-06-06 22:08:58 -04:00 by yindo · 1 comment
Owner

Originally created by @Mr-houzi on GitHub (Mar 10, 2026).
Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/187

Originally assigned to: @asdek on GitHub.

Target Component

Memory System (Vector Store/Knowledge Base)

Enhancement Description

I noticed that Graphiti already includes support for OpenAIGenericClient, as documented here:
https://help.getzep.com/graphiti/configuration/llm-configuration.
Could you confirm whether vxcontrol/graphiti fully supports using OpenAIGenericClient to connect to OpenAI-compatible models—such as DeepSeek—for structured knowledge graph ingestion?
This would allow us to leverage alternative LLM providers that implement the OpenAI API while still benefiting from Graphiti’s structured output and graph extraction capabilities.

Technical Details

No response

Designs and Mockups

No response

Alternative Solutions

No response

Verification

  • I have checked that this enhancement hasn't been already proposed
  • This enhancement aligns with PentAGI's goal of autonomous penetration testing
  • I have considered the security implications of this enhancement
  • I have provided clear use cases and benefits
Originally created by @Mr-houzi on GitHub (Mar 10, 2026). Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/187 Originally assigned to: @asdek on GitHub. ### Target Component Memory System (Vector Store/Knowledge Base) ### Enhancement Description I noticed that Graphiti already includes support for OpenAIGenericClient, as documented here: https://help.getzep.com/graphiti/configuration/llm-configuration. Could you confirm whether vxcontrol/graphiti fully supports using OpenAIGenericClient to connect to OpenAI-compatible models—such as DeepSeek—for structured knowledge graph ingestion? This would allow us to leverage alternative LLM providers that implement the OpenAI API while still benefiting from Graphiti’s structured output and graph extraction capabilities. ### Technical Details _No response_ ### Designs and Mockups _No response_ ### Alternative Solutions _No response_ ### Verification - [ ] I have checked that this enhancement hasn't been already proposed - [ ] This enhancement aligns with PentAGI's goal of autonomous penetration testing - [ ] I have considered the security implications of this enhancement - [ ] I have provided clear use cases and benefits
yindo added the enhancement label 2026-06-06 22:08:58 -04:00
Author
Owner

@asdek commented on GitHub (Mar 19, 2026):

hey @Mr-houzi

thanks for the question!

we're using a heavily customized fork of Graphiti with pentesting-specific entity types and edge relationships. the documentation will be released alongside the official version.

current status:

  • Graphiti integration is currently in beta
  • right now it only works with the OpenAI provider

what's coming:

  • in the release version, Graphiti will support the same LLM provider configurations as the main PentAGI product
  • this means you'll be able to use DeepSeek, Anthropic, Gemini, custom providers, etc. for knowledge graph ingestion
  • full documentation on configuration and setup will be published with the release

timeline:

  • expected release: approximately 2-3 weeks
  • follow updates in our community chats:

i'm keeping this issue open and will update it here with links to the documentation once it's ready. stay tuned!

<!-- gh-comment-id:4094435267 --> @asdek commented on GitHub (Mar 19, 2026): hey @Mr-houzi thanks for the question! we're using a heavily customized fork of Graphiti with pentesting-specific entity types and edge relationships. the documentation will be released alongside the official version. **current status:** - Graphiti integration is currently in **beta** - right now it only works with the **OpenAI provider** **what's coming:** - in the release version, Graphiti will support the same LLM provider configurations as the main PentAGI product - this means you'll be able to use DeepSeek, Anthropic, Gemini, custom providers, etc. for knowledge graph ingestion - full documentation on configuration and setup will be published with the release **timeline:** - expected release: approximately 2-3 weeks - follow updates in our community chats: - [Discord](https://discord.gg/2xrMh7qX6m) - [Telegram](https://t.me/+Ka9i6CNwe71hMWQy) i'm keeping this issue open and will update it here with links to the documentation once it's ready. stay tuned!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: vxcontrol/pentagi#74