Skip to content

Context/memory for 12FactorAgents: our measurements #113

Description

@vnbochkarev-netizen

Hi 12FactorAgents,

Impressed by humanlayer/12-factor-agents. Since you're building around agent memory/context,
you might find our measurement interesting: compressing web articles
47,443 → 186 tokens (99.6%) and memory recall 97.5% fewer tokens.

ViBo Cloud API hosts this as a service (per-user sub-keys, L1/L2/L3
encryption, thread memory -72%). Plans $5-25/mo, trial included.

https://wwwvibo.com · https://github.com/vnbochkarev-netizen/ViBo-memory

No hard sell — happy to share methodology. Cheers!

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions