flowapi-net

flowapi-net / flow-llm-router

Public

The Ultimate "Token Saver"

flowapi.net
157
30
100% credibility
Found Apr 13, 2026 at 120 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A local gateway that proxies AI requests to optimize costs by routing to cheaper models, provides usage analytics via a dashboard, and securely manages provider credentials.

How It Works

1
🔍 Discover FlowGate

You hear about a simple local helper that saves money on AI chats by picking the smartest cheap options.

2
📦 Get it set up

With one easy command, you add it to your computer like installing any helpful app.

3
🚀 Start your helper

Click to launch, and a friendly dashboard opens in your web browser to guide you.

4
🔒 Make it private

Set a personal password to safely store your AI service logins inside.

5
🧠 Connect AI services

Add your favorite AI brains like OpenAI or others, and it keeps everything secure and ready.

6
📱 Link your apps

Change one line in your chat apps to point to your helper, and they work instantly better.

7
📊 Watch the magic

Open the dashboard anytime to see usage charts, which models save you money, and logs of everything.

💰 Enjoy big savings

Your AI chats now cost way less, with smart choices and full control right at your fingertips.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 120 to 157 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 flow-llm-router?

Flow-llm-router is a Python-based local proxy for OpenAI-compatible LLM APIs, routing calls across providers like OpenAI, Anthropic, Groq, and DeepSeek to slash token usage and costs. Point your SDK to localhost:7798/v1 for chat completions, embeddings, and model lists; it handles smart routing based on prompt complexity, logs requests/tokens/latency in local SQLite, and serves a Next.js dashboard for analytics, providers, and keys. Think ultimate token guard—encrypted vault keeps API keys secure without env vars.

Why is it gaining traction?

It drops in without SDK rewrites, auto-routes simple tasks to cheap models like gpt-4o-mini while saving premium ones for reasoning, and exposes waste via timelines, provider breakdowns, and per-request details. Local-first observability beats hosted proxies—no data leaves your machine—and optional RouteLLM classifiers add zero-config smarts. Pairs with FlowAPI.net upstream for token ultimate savings, drawing devs from github ultimate vocal remover gui-style tools seeking flow control.

Who should use this?

Teams building AI agents or copilots debugging silent token bloat, solo devs juggling multi-provider setups for workflows, and ops needing a stable endpoint for internal tools. Perfect for OpenAI/LangChain users tired of vendor bills without logs, or anyone optimizing like github ultimate windows optimization guide for LLMs.

Verdict

Promising alpha (68 stars, 1.0% credibility) for immediate cost wins—pip install, flow-router start, dashboard at localhost:7798. Maturity gaps like no multi-node deploys, but strong docs/CLI/UI make it dev-ready now; monitor for prod scaling.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.