open-world-project

open-world-project / model-router

Public

Automatic cost-aware model routing plugin for Hermes Agent

ai-agent hermes-agent llm-routing openrouter plugin
15
2
100% credibility
Found May 11, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A plugin for the Hermes AI agent that automatically selects cost-effective AI models for each conversation turn based on task complexity across five tiers.

How It Works

1
🧠 Discover smart AI routing

You learn about a helpful add-on for your Hermes AI assistant that picks the perfect AI brain for each task to save money and boost smarts.

2
📥 Add the helper

Download it into your AI assistant's special add-ons folder and run the easy one-click setup.

3
⚙️ Pick your AI favorites

Choose five levels of AI helpers from quick and cheap to powerful thinkers in a simple list.

4
▶️ Start your assistant

Launch Hermes, and it checks everything is set up perfectly before you begin.

5
💬 Chat away

Type your questions or tasks naturally, and it automatically selects the best AI for the job.

6
🔄 Watch it adapt

Notice it switch to stronger AIs for tough problems or let you lock in a favorite with a quick note.

💰 Smarter and cheaper

Your AI tackles everything efficiently, saving you money while delivering top results every time.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is model-router?

Model-router is a Python plugin for the Hermes Agent that automatically routes LLM calls across five cost-optimized model tiers, classifying each user turn by complexity to pick the cheapest capable model—like fast triage on Qwen Flash (T1) or deep reasoning on Claude Sonnet (T5). It solves the problem of ballooning API costs in multi-model agent workflows by escalating only on tool errors or high-stakes tasks, with seamless de-escalation after. Users get per-turn automation, slash commands like /t1-/t5 for pinning, and profile-specific YAML configs, all via a simple git clone and install.sh.

Why is it gaining traction?

Unlike manual model switching or static configs, model-router delivers hands-off cost savings with explicit hints (T3 in chat), auto-escalation on failures, and WebUI integration showing live tiers in the composer dropdown. Developers love the status bar [Tx] badges, startup validation that auto-repairs Hermes files, and support for OpenRouter or Azure OpenAI via model router LLM setups. It's a smart agent upgrade that feels invisible until it saves money on daily runs.

Who should use this?

Hermes Agent powerusers juggling profiles for coding, SEO, or trading who burn through tokens on overkill models. AI workflow builders tired of picking models per task, especially with automatic routing for agent chains. Teams on OpenRouter or Azure wanting model router github automation without constant oversight.

Verdict

Grab it if you're deep in Hermes—install.sh handles everything, docs are solid, but with 15 stars and 1.0% credibility score, treat it as experimental for production. Solid for cost-aware prototyping, just watch for Hermes updates breaking patches.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.