Wistom Zhang 1fbc06279a
feat(benchmark): add offline memory vs project-context comparison suite (part of #7884) (#8058)
* feat(benchmark): add offline memory vs project-context comparison suite (part of #7884)

* fix(benchmark): route prompts through DynamicContextMiddleware, flatten rows, and share test run

* refactor(benchmark): run through local Gateway and simplify first-pass suite (#7884)

* docs(agents): keep context memory benchmark guidance within chain budget

* fix(benchmark): verify DeerMem scope rejection

---------

Co-authored-by: Totoro <msy626836554@gmail.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
2026-11-10 14:45:56 +08:00

29 lines
749 B
JSON

{
"dataset": "context-memory-synthetic-v1",
"question": "Which context applies here?",
"thread_history_control": "BENCH-THREAD-HISTORY",
"default_fact": "BENCH-DEFAULT-PREFERENCE",
"agent": "bench-agent",
"agent_fact": "BENCH-AGENT-PREFERENCE",
"project_only_fact": "BENCH-PROJECT-ONLY-PREFERENCE",
"projects": {
"alpha": {
"instructions": "BENCH-ALPHA-INSTRUCTIONS",
"document": "BENCH-ALPHA-DOCUMENT.md"
},
"beta": {
"instructions": "BENCH-BETA-INSTRUCTIONS",
"document": "BENCH-BETA-DOCUMENT.md"
}
},
"invalidation": {
"replacement": {
"old": "BENCH-OLD-REPLACEMENT",
"new": "BENCH-NEW-REPLACEMENT"
},
"deletion": {
"old": "BENCH-OLD-DELETION"
}
}
}