Show HN: Llmspeed – feel how fast a local LLM runs before you buy the hardware
Category: devtools
Tags: llm, benchmarking, hardware, performance, simulation, local-ai
Score: 7.0/10 (Innovation: 7, Technical: 6, Documentation: 8, Utility: 7)
Llmspeed is a web tool that simulates the real-time performance of local LLMs on various hardware configurations, including time to first token and tokens per second. It uniquely combines measured benchmarks with first-principles modeling to provide estimates even for hardware without existing benchmarks, helping users decide on hardware purchases for running LLMs locally.
Target audience: AI practitioners, hardware enthusiasts, developers running local LLMs
Repository: https://llmspeed.dev/ · TypeScript · MIT · 150 stars
View on Hacker News