I built an agent Harness for Small Models. I got Qwen 3.5 4b managing servers.
Summary
Built an agent harness for small models, enabling Qwen 3.5 4b to manage servers.
Similar Articles
Built an agent harness with cost control and native MCP support
Agentic Harness is a new open-source Python infrastructure that turns YAML files into executable agent workflows, featuring markdown audit logs, pre-built specialists, cost guardrails, native MCP support, and local model integration.
I used local Qwen 27b to build a harness and replace OpenCode
The author built an open-source harness for running local LLMs using Qwen 27b, featuring just-in-time code review, sub-agents, voice dictation, and more, and shares development insights.
Remote agent harness
A tool for remotely harnessing and managing AI agents.
What actually breaks agents on a 4B model? Notes from 50 runs on Qwen3.5-4B
A practitioner ran 10 agent harnesses on Qwen3.5-4B across 5 tasks and found that failures came from exact-match edits, literal instruction-following, miscounting, and API mismatches — not malformed tool calls. Six of ten agents passed all tasks, with an 11x spread in run time on identical hardware.
My agent kept losing track of itself between sessions, so I rebuilt the harness instead of switching models
The author found that AI agent reliability issues were caused by the harness rather than the model, and improved it by separating context tracking, loading, and output checking, while working on versioning to manage code across projects.