Aifloxium
Aifloxium

Complete Setup Guide: $0 Agentic Coding Stack (North Mini Code & GLM-5.2)

2026-06-27
Muhammad Shadab Shams
AI Automation

"A step-by-step setup guide to deploying a production-ready developer agent stack for $0. Learn how to configure Cohere North Mini Code and GLM-5.2 with OpenRouter and Cloudflare Workers AI for developer frameworks."

Complete Setup Guide: $0 Agentic Coding Stack (North Mini Code & GLM-5.2)
Executive Summary // TL;DR

Setting up a production-ready agentic coding workspace does not require expensive closed-source model subscriptions. By combining OpenRouter's free tier (hosting Cohere North Mini Code) with Cloudflare Workers AI (hosting GLM-5.2), you can build a fully autonomous, zero-cost developer stack. This guide walks you through step-by-step credentials configuration and client launch.

Model Stack Specifications

Swipe to Explore
FeatureNorth Mini Code (OpenRouter Free)GLM-5.2 (Cloudflare Workers AI)
Architecture30B MoE (3B active)Dense Coding Model
Context Window32k tokens262k tokens
Native FeaturesAgentic coding loopsNative Tool Calling
Daily LimitModel dependent (high)10,000 neurons/day ($0)

01

Method 1: Cohere North Mini Code

Free via OpenRouter

North Mini Code is a 30B Mixture-of-Experts (MoE) model with only 3B active parameters, specifically tuned for multi-step agentic coding tasks.

1. Create an OpenRouter Account

  • Go to OpenRouter.ai.
  • Sign up or log in (no credit card or billing details required for free models).

2. Generate an API Key

  • Navigate to Keys (or Account Settings -> API Keys).
  • Click Create Key and give it a name (e.g., Coding Agents).
  • Copy the generated key immediately (it looks like sk-or-v1-...).

3. Configure Environment Variables

For macOS / Linux (Bash):

bash
1export OPENROUTER_API_KEY=your_copied_api_key_here

For Windows (PowerShell):

powershell
1$env:OPENROUTER_API_KEY="your_copied_api_key_here"

Tip: Add this command to your shell profile (~/.bashrc, ~/.zshrc) to make the variable persistent across sessions.

4. Add to Agent Configuration

Configure your developer agent (e.g., OpenCode, Hermes) to use the following provider credentials:

  • Base URL: https://openrouter.ai/api/v1
  • Model ID: cohere/north-mini-code:free

02

Method 2: GLM-5.2 Configuration

Free via Cloudflare Workers AI

GLM-5.2 is a state-of-the-art coding model featuring native tool calling capabilities and a massive 262,144 token context window, running 100% free with a daily allocation of 10,000 neurons (resets daily).

1. Create a Cloudflare Account

  • Sign up for a free account at dash.cloudflare.com.
  • No paid subscriptions are required for the daily free Neurons allocation.

2. Locate your Account ID & REST API

  1. From the left-hand navigation sidebar, toggle the AI menu and select Workers AI.
  2. On the Workers AI dashboard homepage, click the REST API button to load details.
  3. Locate and copy your alphanumeric Account ID displayed in the panel.

3. Create an API Token

  1. Go to My Profile (user icon in top-right menu -> click My Profile).
  2. Click API Tokens in the sidebar -> click Create Token.
  3. Choose Create Custom Token (at the bottom).
  4. Configure Custom Token permissions: Account -> Workers AI -> Edit.
  5. Click Continue to summary and confirm with Create Token.
  6. Copy the generated custom token.

4. Configure custom provider details

Add Cloudflare as a custom OpenAI-compatible endpoint in your agent configuration:

  • Base URL: https://api.cloudflare.com/client/v4/accounts/YOUR_ACCOUNT_ID/ai/v1
  • Header Configuration: Authorization: Bearer YOUR_API_TOKEN
  • Model ID: @cf/zai-org/glm-5.2

03

Start Your Agent

Execution

Once your credentials are set, launch your coding agent environment (such as opencode or hermes-agent) pointing to the model you configured.

For example, to execute a local CLI developer agent using your OpenRouter free model parameters, run:

bash
1hermes-agent --model cohere/north-mini-code:free --api-key $OPENROUTER_API_KEY

You now have a fully operational, zero-cost developer agent capable of reading file systems, performing search actions, executing CLI commands, and editing codebases.

The Directive

Looking for advanced agent structures?

To explore persistent SQLite memories, telegram control loops, and custom terminal setups, read my comprehensive Hermes Agent Setup Guide.


Stack FAQs

Scale Your AI Infrastructure.

Ready to transition your workflows to multi-agent automation? Contact me today for a custom implementation audit.

Phone

+923464883396

Primary Email

info@aifloxium.online

Direct Email

muhammadshadabshams@gmail.com

Website

www.aifloxium.online

You will speak directly with Muhammad Shadab Shams. Best fit: teams seeking automated workflows, custom internal operations tools, or AI integration. Get a free custom automation flowchart of your current workflow during our call.

No spam. Scoping response within 24 hours.