Monarchic LLC logo Monarchic LLC

Product Research

Benchmarks
With Receipts

These pages show the tests behind the products: what ran, what passed, and where the results stop. The point is not to make every score sound impressive. It is to make the evidence readable.

Memory Runtime

ExplicitMem

97.8%

LongMemEval-S accuracy

ExplicitMem is a private memory MCP runtime for typed memory objects, graph links, bounded context synthesis, and receipt-backed retrieval.

Repository Intelligence

RepoIntel

0.9979

RepoIntel-Bench score

RepoIntel is a repository intelligence MCP runtime for indexing code, symbols, evidence records, generated wiki pages, and change-aware retrieval.

Browser Verification

BrowserOps

7/7

Reliability checks passed

BrowserOps is a browser evidence and UI verification MCP for agents that need screenshots, console and network diagnostics, scenario checks, and receipt-backed browser QA.

Scope

These notes are scoped to specific benchmarks. A score can show progress inside a harness, but it does not prove every real-world use case. Product pages should change when datasets, scoring rules, or runtime boundaries change.