Skip to content

Conversation

Ololoshechkin
Copy link
Collaborator

@Ololoshechkin Ololoshechkin commented Jun 17, 2025

Fixes #288

Allows to select an optional custom model for fact retrieval from the history in memory nodes and history compression nodes


Type of the change

  • New feature
  • Bug fix
  • Documentation fix

Checklist for all pull requests

  • The pull request has a description of the proposed change
  • I read the Contributing Guidelines before opening the pull request
  • The pull request uses develop as the base branch
  • Tests for the changes have been added
  • All new and existing tests passed
Additional steps for pull requests adding a new feature
  • An issue describing the proposed change exists
  • The pull request includes a link to the issue
  • The change was discussed and approved in the issue
  • Docs have been added / updated

@Ololoshechkin Ololoshechkin requested a review from sdubov June 17, 2025 01:35
Copy link

github-actions bot commented Jun 17, 2025

Qodana for JVM

377 new problems were found

Inspection name Severity Problems
Check Kotlin and Java source code coverage 🔶 Warning 278
Missing KDoc for public API declaration 🔶 Warning 77
Unused import directive 🔶 Warning 18
Vulnerable imported dependency 🔶 Warning 4
@@ Code coverage @@
+ 64% total lines covered
7573 lines analyzed, 4917 lines covered
# Calculated according to the filters of your coverage tool

☁️ View the detailed Qodana report

Contact Qodana team

Contact us at [email protected]

Copy link
Collaborator

@tiginamaria tiginamaria left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM in general, just one concern about the way messages cleanup is implemented

Copy link
Contributor

@sdubov sdubov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Besides comment from @tiginamaria , the change LGTM!

@aozherelyeva
Copy link
Contributor

The PR is not linked to the issue that it fixes (only a commit is mentioned there), so here is the link: #288

@Ololoshechkin Ololoshechkin force-pushed the vbr/memory-model-selection branch from 84ebea6 to a045616 Compare July 12, 2025 15:37
@kpavlov
Copy link
Collaborator

kpavlov commented Jul 12, 2025

Pls, update the description :)

Copy link
Collaborator

@kpavlov kpavlov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you, @Ololoshechkin
Just one suggestion

*/
public suspend fun saveFactsFromHistory(
concept: Concept,
subject: MemorySubject,
scope: MemoryScope,
retrievalModel: LLModel? = null
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we make retrieval model required?

Suggested change
retrievalModel: LLModel? = null
retrievalModel: LLModel = model

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No, it won't work, unfortunately, because model is only available inside llm.writeSession as it's a subject to change throughout the graph execution.

@Ololoshechkin Ololoshechkin changed the title #288: Support providing custom LLM for fact retrieval in the memory #288: Support providing custom LLM for fact retrieval in the history Jul 12, 2025
@Ololoshechkin Ololoshechkin merged commit a733113 into develop Jul 12, 2025
5 checks passed
@Ololoshechkin Ololoshechkin deleted the vbr/memory-model-selection branch July 12, 2025 22:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Support selection of a model in memory feature
5 participants