Recallium vs Supermemory

Feature comparison

Product details checked September 8, 2026; features and availability vary by edition. The benchmark figures measure different things: Recallium’s headline is retrieval recall at depth ten and Supermemory’s at depth fifteen “with aggregation”, a term their report never defined, so this is not a controlled head-to-head ranking. Supermemory’s recall@15 was published at supermemory.ai/research/longmembench, which now redirects to their home page; no archived copy exists, so the link above goes to their current research index. Sources: Supermemory documentation and GitHub.

How they differ in practice

Different product categories, benchmarked on common ground. Supermemory is a memory and context engine for applications: it ingests documents and connected services alongside conversation memories, scoped by container tags. Recallium provides shared team context for coding agents: it captures what agents learn while working in a repository, architecture, decisions and fixes, and gives it to every agent and teammate on that codebase. Both are measured on LongMemEval-S because retrieval quality is measurable and comparable, not because the products are substitutes.

When to choose Recallium

• Your agents work in a shared codebase and need the same context
• You want decisions, patterns and fixes captured from agent sessions, not just facts about users
• Your team uses Claude, Cursor, Codex or VS Code and wants one memory across them
• You want a benchmark whose reader, judge and depth are published

When to choose Supermemory

• You need general-purpose memory for an application, not a coding workflow
• You want connectors that pull documents and services into memory
• You want to run it locally from the open-source repository today