Evals
Evals
Browse the engine and local-model measurement scoreboard.
How it works
Open Evals to browse measured engine and model results. Read the workload and measurement notes with each result to understand what it covers.
Related research
- A protocol for comparing engine performance
See how the study matched inputs and output checks and kept each timing boundary explicit.
- Repeating CSV measurements across sessions
See which results held up when the benchmark was rerun and where uncertainty remained.
- Measuring evaluation throughput and latency
See why the study measures a whole batch separately from the time one chain takes.
Example
- Start with
Comparing data engines
- Try this
Open the Evals scoreboard
- Result
Inspect the matching workload