Show HN: RightSize CLI, Find the cheapest LLM that works for your prompt
Posted 8 hours ago by 44za12
3 points
https://github.com/NehmeAILabs/rightsize-cliRightSize runs your prompt against candidate models (Kimi, GLM, Qwen, Gemma etc.) in parallel via OpenRouter. Then it uses a stronger model as a Judge to score accuracy against a baseline model.