Home / LLM Inference / GroqGGroqFastest LLM inference on LPUsReviewed byRoman KAboutHardware-accelerated inference for Llama, Mixtral, and more. Famous for hundreds of tokens/sec.inferencelpufastopenai-compatibleMetricsMore in LLM InferenceTogether AIRun open-source models in productionFireworks AIProduction inference for open modelsOpenRouterUnified API for hundreds of LLMsModalServerless GPUs and containers