WORKSPACE MEMORY / BENCHMARKS
Compare memory across agents.
One prompt. Shared context. Every memory provider.
Sample data
Compare in detail
vs
Expected answer
Across all providers
Current task · every memory provider
Time to answer
Input token use
Run review
Checks, answers, and tool trace in one view.
Metrics by agent
Every agent for the two selected providers.
| Agent | Provider | Input | Output | Tools | Latency | Cost | Result |
|---|
Same prompt and source snapshot across every run.1 run / agent / provider