Offline LLM Inference Chat

Offline, Large Language Model chat capability for Mobile Games.

  • Supported Platforms
  • Supported Engine Versions
    5.3 - 5.4
  • Download Type
    Engine Plugin
    This product contains a code plugin, complete with pre-built binaries and all its source code that integrates with Unreal Engine, which can be installed to an engine version of your choice then enabled on a per-project basis.

The LLM inference Plugin provides simplified access to offline Large Language Model chat capabilities for use in mobile games.


It includes samples and guides on building your own purpose driven A.I. chatbot for in game characters and general chat. Also, included are guides on how to use the extensive and growing list of open source LLM models and LoRas. The plugin is simple to use and suitable for beginners of Large Language models.


The Plugin implements a GameInstance Subsystem that provides configuration options to an on device LLM inference model runtime. 



Technical Details

Features:

  •  Offline LLM chat
  •  Compatible with GEMMA_2B, PHI_2, FALCON_RW_1B, STABLELM_4E1T_3B
  •  Guides and Tutorials

Code Modules:

  •  LLMInference - Runtime

Number of Blueprints: 7

Number of C++ Classes: 5

Network Replicated: NO

Supported Development Platforms: Windows

Supported Target Build Platforms: Android

Documentation: Docs

Example Project:

Important/Additional Notes: