JoshXT/AGiXT-Qwen3-4B
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Jan 30, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm
JoshXT/AGiXT-Qwen3-4B is a 4 billion parameter text-based causal language model, fine-tuned from Qwen3-4B-Instruct-2507 by JoshXT. It is specifically optimized for AGiXT agent interactions, natively understanding AGiXT's XML-based command syntax, tool delegation patterns, and structured response formats. This model excels at reasoning through problems and executing commands within the AGiXT ecosystem.
Loading preview...
AGiXT-Qwen3-4B: Core Text Model for Intelligent Agents
This 4 billion parameter text model, fine-tuned by JoshXT from Qwen3-4B-Instruct-2507, is a core component of the AGiXT agent ecosystem. It has been extensively trained on a specialized Agent Interaction Dataset to deeply understand and generate AGiXT's unique operational patterns.
Key Capabilities
- Native AGiXT Command Understanding: Processes and generates AGiXT's XML-based command execution format, including
<execute>,<name>, and<param>tags. - Structured Reasoning: Utilizes
<thinking>blocks for internal reasoning before command execution and provides consistent<answer>formatting. - Intelligent Tool Delegation: Knows when to delegate tasks, such as coding to GitHub Copilot, or using other built-in AGiXT extensions like
web_browsingorpostgres_database. - Extension Awareness: Understands the capabilities and proper parameter usage for over 778 AGiXT commands across various extensions.
Good For
- AGiXT Agent Workflows: Ideal for agents requiring precise command generation and structured responses within the AGiXT framework.
- Complex Text-Based Tasks: Handles multi-step reasoning and command execution for tasks that don't require visual input.
- Integration with AGiXT's Routing System: Designed to work seamlessly with the
AGiXT-AbilitySelect-270mmodel for efficient, complexity-based task routing.