WikiofAuto
+41.2% Valid n8n typeVersion & Connection Wiring Accuracy vs GPT-4oMethod: QLoRA Fine-Tune on 5,778 Verified Current n8n v1.x Workflows & FastAPI/Astro ReposSize: 32.5B

FlowCoder-32B (n8n DAG, Python & TypeScript Automation Architect)

Trained on verified current n8n workflow DAGs (agent v3.1, httpRequest v4.2, code v2), Python automation scripts, and Cloudflare Worker relays. Generates importable n8n JSON and JavaScript Code nodes.

BASE ARCHITECTURE
Qwen/Qwen2.5-Coder-32B-Instruct
VRAM REQUIREMENT
18 GB VRAM
QUANTIZATION FORMATS
Q4_K_M (19.8 GB), IQ4_XS (17.6 GB), Q6_K (26.7 GB)
PARAMETERS
32.5B
QUICK DEPLOY COMMAND (OLLAMA / LLAMA.CPP):ollama run qwen2.5-coder:32b

Eliminates Deprecated n8n Node Hallucinations

General-purpose LLMs trained on pre-2025 web data generate deprecated n8n v1 Function nodes and broken connections syntax. FlowCoder-32B is grounded in current n8n v1.x+ schemas (n8n-nodes-base.code v2, @n8n/n8n-nodes-langchain.agent v3.1) and pairs directly with our n8n-forge compiler.