Skip to content

feat(eval): add openmem-eval offline + live benchmarking kit (004)#1

Merged
Ketankhunti merged 1 commit into
mainfrom
004-eval-kit
May 1, 2026
Merged

feat(eval): add openmem-eval offline + live benchmarking kit (004)#1
Ketankhunti merged 1 commit into
mainfrom
004-eval-kit

Conversation

@Ketankhunti

Copy link
Copy Markdown
Owner

Stdlib-only eval CLI with bundled 50-fact/20-query dataset, recall@k/MRR scoring, latency percentiles, JSONL trace, markdown report, cost estimation with confirmation gate, and post-run cleanup. Supports postgres/mem0/supermemory/letta/passthrough providers via OMP_POSTGRES_URL and *_API_KEY envs. 365 tests, 89% coverage. Live postgres run validated: recall@5=0.80, MRR=0.78, search p95=9.7ms.

Stdlib-only eval CLI with bundled 50-fact/20-query dataset, recall@k/MRR scoring, latency percentiles, JSONL trace, markdown report, cost estimation with confirmation gate, and post-run cleanup. Supports postgres/mem0/supermemory/letta/passthrough providers via OMP_POSTGRES_URL and *_API_KEY envs. 365 tests, 89% coverage. Live postgres run validated: recall@5=0.80, MRR=0.78, search p95=9.7ms.
@Ketankhunti
Ketankhunti merged commit d0541e1 into main May 1, 2026
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants