Compare Realistic Agentic Inference Perf

Long Context Multi Turn Inference Performance. Compare Across Google TPUv7 Ironwood, OpenAI Jalapeño, MI355X, GB300 NVL72, GB200 NVL72, B200, H200, H100, RTX Pro, and soon TPUv8i, SambaNova SN50, Rubin NVL72, MI455X UALoE72

Token Revenue Calculator Dashboard

TPU Qwen3.5 SubmissionNEWAgentXView results TPU Kimi K3 SubmissionNEW2.8 Trillion parameters AgentXView results Vera Rubin NVL72 GLM 5.3 Submission744 Billion parameters AgentXView results Vera Rubin NVL72 MiniMax M3 Submission428 Billion parameters AgentXView results OpenAI JalapeñoDeepSeek R1 · 8K/1K · FP4View results

Open-Source Continuous AI Accelerator System Benchmark Trusted by GigaWatt Token Factories

Every Result Is Transparently done through Public GitHub Actions Automation

Every data point on the dashboard is produced by a public GitHub Actions workflow run. The recipe lives in the repo, the run executes on the actual target hardware, and the full logs and artifacts are publicly viewable. Click any point on a chart to jump straight to the run that produced it. All reproducible, auditable, and open source.

1,000+ new benchmark datapoints added per week on average. Browse every new model, chip, framework, and configuration as it lands.

Public Actions runs

Every benchmark executes on GitHub Actions with full logs visible while the run is in progress.

Open recipes

Every model, framework, precision, and parallelism setting is committed to the public repo as a shell script.

Browse submissions View benchmark runs on GitHub Actions How it works