Spark Arena
LLM Performance Leaderboard
Compare Large Language Model performance on NVIDIA DGX Spark infrastructure. Real benchmarks, real hardware, real results.
Real Performance
Benchmarks from actual llama-benchy runs on DGX Spark hardware
Multiple Runtimes
Compare vLLM, SGLang, TensorRT-LLM, and llama.cpp implementations
Full Transparency
View complete recipes, configurations, and detailed benchmark results
Built on Open Source
This leaderboard is powered by community-developed tools. Support the projects that make this possible!
spark-vllm-docker
Docker containers and recipes for running LLM inference engines on NVIDIA DGX Spark
llama-benchy
Comprehensive benchmarking tool for measuring LLM inference performance across different workloads
sparkrun
Launch, manage, and stop LLM inference workloads on NVIDIA DGX Spark systems
Deep Dives, Runbooks, and Benchmarking Notes
Follow Spark Arena engineering write-ups covering reproducibility, runtime tuning, and practical performance investigations from the community.