All activity
aplomb2left a comment
Hey everyone! I'm Bo, an indie app developer running a portfolio of AI-powered apps. Why I built this: Last month I looked at my Claude Code bill ā $212 for one person. I dug into the API logs and realized 70%+ of my tokens were going to simple tasks: reading files, writing commit messages, running grep. All through Opus at $75/M tokens. Meanwhile, Gemini Flash could handle those same tasks for...
CodeRouterCut your AI coding bill 70% with automatic task routing
Claude Code and Cursor route everything through Opus ($75/M tokens) ā even file reads and commit messages. CodeRouter adds phase-aware task routing: it detects what you're doing and picks the optimal model. Planning gets Opus. Implementation gets Sonnet/DeepSeek. Tests get Flash. Real result: $212/mo to $63/mo. One URL change, same workflow. Works with Cursor, Claude Code, Aider, Copilot, Windsurf. 1M free tokens on signup.
CodeRouterCut your AI coding bill 70% with automatic task routing
Stop overpaying for AI API calls. ClawRouters analyzes each query and auto-routes to the optimal model.
Simple task? ā Cheap model. Complex reasoning? ā Claude or GPT-4o.
One API endpoint. OpenAI-compatible. Drop-in replacement.
š Smart routing across Claude, GPT-4o, Gemini, DeepSeek
š° 40-60% cost savings, no quality loss
š One API key for all providers
š Real-time usage dashboard
š§ BYOK supported
Built for developers tired of overpaying or manually switching models.
ClawRoutersOne API that auto-picks the best LLM. Save 40-60%.
aplomb2left a comment
Hey Product Hunt! š I built ClawRouters because I was spending $200+/month on GPT-4o API calls, and realized ~70% of my queries were simple enough for much cheaper models. But manually switching between providers was a nightmare ā different SDKs, API formats, rate limits... So I built one endpoint that figures it out for you. Send your prompt, ClawRouters analyzes the complexity, picks the...
ClawRoutersOne API that auto-picks the best LLM. Save 40-60%.