Skip to content

v2.3.0 - Testing Release (Single-Model Agent)

Latest
Latest

Choose a tag to compare

View all tags
@hussainnazary2 hussainnazary2 released this 11 Sep 15:58
· 2 commits to main since this release
v2.3.0
490e0e6

Testing Release (Single-Model Agent)

This is a testing/preview release locked to one model (Gemma 4 12B Instruct Q4_K_M) while we validate the new plan-driven agent architecture.

v2.3.0 - Chat Interface

v2.3.0 - Agent Mode

v2.3.0 - Settings

What's new in v2.3.0

  • Single-model focus (temporary) - optimized for Gemma 4 12B Q4_K_M; multi-model detection removed so this one model just works
  • Strictly plan-driven agent - every turn goes through a planner node: tool-free questions answered directly; tasks get a step-by-step plan executed with sandboxed tools
  • Developer-style process UI - plan steps, tool calls, and results render inline in the chat above each answer
  • Reliable final answers - stray tool-call JSON and stale status text are scrubbed from replies
  • Auto-load at startup - scans models/ folder and loads the pinned GGUF automatically; missing model downloads with live progress
  • Pruned tools - 10 focused workspace tools (memory/meta tools removed)
  • Self-setup launchers - launch.bat / launch.sh handle all dependency installation

Full Capabilities

Agent Tools:

Tool What it does
list_directory / glob Explore folders and match file paths in the workspace
read_file Read any file (MD/PDF/DOCX/TXT/code - text extracted automatically)
search_files Find files and grep for content
write_file / edit_file / move_file Create, edit, and move files
run_command Run a shell command inside the workspace (approval-gated)
run_python Execute Python source inside the workspace (approval-gated)
git Git operations - read-only runs freely, writes require approval

Advanced Search (Find Paragraph, no RAG):

  • Single file - type a question and the model finds and ranks matching paragraphs
  • Folder search - a planner decides which files to read, with live per-file progress
  • Smart defaults - your last query, source, and settings are remembered between sessions

Interface Features:

  • Sessions left, chat center, tools right layout
  • Streaming chat via WebSocket
  • Model chip in header with live download progress
  • Settings dialog (Model, Providers, Agent, Hardware, Appearance, Keyboard, Plugins)
  • Dark/light theme with accent colors
  • Command palette (Ctrl/Cmd+K)
  • Electron standalone desktop app

Core Features:

  • Plan-driven agent (LangGraph) with Allow/Deny approval cards
  • SQLite checkpointing - conversations survive restarts
  • Real file reading (MD/PDF/DOCX/TXT/code)
  • GPU acceleration via Settings -> Hardware
  • 100% local inference - no data leaves your machine
  • Cross-platform (Windows 10/11, Linux, macOS)

Downloads

Artifact Size Notes
GGUFLoader_v2.3.0_CPU.exe ~145 MB Windows - CPU-only, works everywhere
GGUFLoader_v2.3.0_GPU.exe ~930 MB Windows - NVIDIA CUDA
GGUFLoader_v2.3.0_linux_x86_64_CPU ~50 MB Linux - CPU-only

First launch

  1. Start the app - it auto-loads the pinned Gemma 4 12B Q4_K_M from the models/ folder
  2. No model on disk? The model chip in the header downloads it with live progress
  3. Chat in the main window, or press Ctrl+Shift+A for Agent Mode and choose a workspace folder

System Requirements

  • OS: Windows 10/11, Linux, macOS (Intel & Apple Silicon)
  • RAM: 32 GB minimum recommended
  • Storage: ~8 GB free for the model file
  • GPU: Optional (NVIDIA CUDA on Windows/Linux, Metal on macOS)

Coming next

Universal Model Loader returns - the next release removes the single-model restriction so you can run any GGUF model with the agent, with hardware-aware recommendations.

Full README: https://github.com/GGUFloader/gguf-loader/blob/main/README.md

Assets 5
Loading