ComputeArena
Community-submitted throughput benchmarks for local AI models on edge silicon.
ComputeArena provides community-submitted throughput benchmarks for local AI models running on edge silicon. It allows users to decode and compare performance metrics like tokens per second, based on model, quantization, and chip. The platform uses BaseRT and llama.cpp runtimes for accurate, community-driven hardware performance data.
Free
How to use ComputeArena?
Install the ComputeArena CLI for macOS or Linux. Choose your preferred runtime (BaseRT or llama.cpp). Run benchmarks offline without an account. Once satisfied, sign in to submit your signed reports to share your hardware's performance publicly with the community.
ComputeArena 's Core Features
ComputeArena 's Use Cases
ComputeArena 's FAQ
Most impacted jobs
AI Researcher
ML Engineer
Hardware Enthusiast
Developer
Data Scientist
ComputeArena 's Tags
#AI Benchmarking#Local LLM#Edge AI#Performance Metrics#Hardware Comparison#Tokens per Second#Quantization#CLI Tool
