Loom is tested on three key AI memory benchmarks: one for consistency as facts change, one for recall across long histories, and one for memory over real agent work. Together they ask whether an agent can recover the right current context when it matters.
Changing-history memory. Published prior SOTA comparator: CUPMem at 78.7%.
Long-horizon memory QA. Published full-context comparator: GPT-4o at 60.6%.
Context: Some vendors publish results in the mid-to-high 90s, often after heavy benchmark-specific optimization. Loom did not engage in benchmark-specific tuning; this score reflects its general memory system.
Memory over realistic agent work. Published full-context comparator: GPT-5.2 at 72.3%.
The table keeps the homepage figures in one place and shows the public alternatives cited by the corresponding benchmark papers or public benchmark writeups.
| Benchmark | Loom | Published alternatives | What the benchmark stresses |
|---|---|---|---|
| STALE | 86.1% | CUPMem 78.7%; GPT-4o-mini without memory 8.7%; Mem0 8.3%; LiCoMemory 7.6%; Zep 6.0%. | Whether memory stays current as facts change, including stale premises and cascading conflicts. |
| LongMemEval | 89.3% | GPT-4o full-context 60.6%. | Long-horizon questions over human history, including updates and temporal reasoning. |
| AMA-Bench | 75.6% | GPT-5.2 full-context 72.3%; AMA-Agent 57.2%; MemoRAG 46.1%. | Whether agents can use memory from prior tool use, task steps, preferences, and outcomes. |
Loom figures are from complete benchmark evaluations. Comparison numbers are shown as published alternatives, not re-run claims for those systems.
A single retrieval score cannot tell the whole story. These benchmarks cover three failure modes an agent encounters in practice: believing something obsolete, losing facts in a long history, and forgetting what happened while another agent was working.
Each scenario contains an earlier fact and a later update that makes it obsolete. The benchmark checks whether the agent identifies the current state, rejects questions built on stale premises, and adapts its response to the new reality.
Questions draw on long, multi-session human histories. Success requires more than finding a sentence: the agent must connect evidence across sessions, order events in time, and honor later knowledge updates.
Instead of human conversation, AMA-Bench uses agent trajectories: actions, tool outputs, observations, state changes, and outcomes. It measures recall, causal inference, state updating, and state abstraction across six domains.
Each evaluation covers the benchmark's full published suite, preserving the complete mix of conflict types, question categories, agent domains, and memory skills it was designed to test.
Loom answers from the structured memory it built during ingestion; it does not receive the original conversation or trajectory at question time. Comparison scores are published reference points from the benchmark papers, not results re-run by Loom.
Connect a hosted store, ask your agent to describe it, then use Loom in the coding workflow you already have open.
Request access Open install guide