HN Showcase
Back to browse
GitHub Repository

Open-source LLM router & AI cost optimizer. Routes simple prompts to cheap/local models, complex ones to premium — automatically. Drop-in OpenAI-compatible proxy for Claude Code, Codex, Cursor, OpenClaw. Saves 40-70% on AI API costs. Self-hosted, no middleman.

515 starsPython

NadirClaw – Open-source LLM router with 10ms classification

by amirdor·Feb 16, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidSolve My ProblemBig Brain

Smart LLM routing cuts costs, but competing against established OpenRouter and vLLM ecosystems.

Strengths
  • •10ms prompt classification with sentence embeddings + agentic/reasoning task detection is genuinely clever; session pinning and context-window fallback solve real multi-turn routing friction.
  • •OpenAI-compatible proxy means zero integration friction — works with Cursor, Continue, Claude Code, plain curl without rewriting applications.
Weaknesses
  • •Routing accuracy depends entirely on embedding-based classification; no evidence of accuracy benchmarks or failure modes vs. OpenRouter's established pricing models.
  • •Python-only, self-hosted deployment model adds operational burden compared to SaaS alternatives with API keys.
Target Audience

Developers, AI engineers, cost-conscious LLM users, teams using OpenAI-compatible tools

Similar To

OpenRouter · vLLM · LiteLLM

Similar Projects

Developer Tools●●Solid

NadirClaw, LLM router that cuts costs by routing prompts right

If you're burning through Claude/OpenAI credits, this is a low-friction stopgap: it classifies prompts in ~10ms and routes trivial tasks to cheaper/local models while reserving premium APIs for complex work. The agentic-task detection, reasoning-aware routing, session pinning and context-window fallback are practical touches that avoid mid-thread model bouncing and 429 failures. It isn't reinventing the space (OpenRouter and others exist), but it's focused on real-world cost tradeoffs and drop-in compatibility.

Solve My ProblemNiche Gem
amirdor
113mo ago
Developer Tools●●Solid

API router that picks the cheapest model that fits each query

Komilion turns model sprawl into a cost-control layer you drop in by swapping a base_url: requests are classified (regex fast path + tiny LLM) and matched to ~400 models so cheap models handle the easy stuff and premium models only run when needed. The ~60% zero‑call regex fast path and benchmark-driven routing (LMArena) are clever, pragmatic moves; the hard questions left are model-quality drift across providers and how routing decisions map to real-world user satisfaction.

Solve My ProblemWizardrySlick
robinbanner
113mo ago