MatrixArk Contact

Context memory benchmarks

Measured on the memory benchmarks the field actually uses.

MatrixArk's context engine is built on TemporalStore. We benchmark it head-to-head against an OpenViking-style direct-source-retrieval baseline on LOCOMO and LongMemEval using one shared open-source stack — the same qwen2.5:7b reader and MiniLM embeddings for both, scored by a Claude LLM judge — so the comparison is about the memory layer, not the model around it.

Head-to-head

TemporalStore ties OpenViking on short chats, and decisively beats it on long-horizon memory.

Head-to-head against an OpenViking-style direct-source-retrieval baseline on the identical OSS stack — same qwen2.5:7b reader, same embeddings, same retrieval budget, Claude-judged. Only the memory layer differs.

BenchmarkMetricTemporalStoreOpenViking (same 7B model)
OverallAnswer accuracy (Claude judge)88%78%
LongMemEvalAnswer accuracy98%66%
LOCOMO (tie)Answer accuracy83%84%
LongMemEvalRetrieval hit@k1.000.88
LongMemEvalPrompt tokens at equal quality97% fewer vs replaying full history

The gap is widest on LongMemEval — long-horizon, multi-session questions — exactly where time-valid, replayable memory with stale-context blocking pays off. On shorter LOCOMO conversations the two are even.

The landscape

How the context approaches differ.

SystemApproachTemporal & replayOne service
MatrixArk / TemporalStoreAppend-structured temporal store; Context modelNative event-time + replayYes — memory, retrieval, counters, replay
Leading hierarchical memoryEvent/entity memory, L0/L1/L2 layersTemporal layersMemory-focused
Mem0Add/search API, scoped memoriesLimitedNeeds a vector store
Zep / GraphitiBi-temporal knowledge graphBi-temporal graphGraph + retrieval
MemOS / MemoriOS/filesystem-style memoryFile-likeMemory-focused

Method

Fair by construction.

Both systems run behind the same open-source harness: identical qwen2.5:7b reader and all-MiniLM-L6-v2 embeddings, the same LOCOMO (multi-session conversation QA) and LongMemEval (long-horizon memory) datasets, and a Claude LLM judge applied uniformly. Only the memory layer differs, so the results isolate the contribution of the store — ingestion, retrieval, freshness, and context-pack assembly.

The enterprise story adds the parts a benchmark cannot show on a laptop: shared-storage throughput under concurrency, five-nines availability, and replayable audit — all on the same engine.

Tokens & quality

A small managed pack beats replaying the whole history — on tokens and on answers.

Three arms over the same questions and the same reader/judge model: replay the full local context, or retrieve a token-budgeted ContextPack. Graded against query-relevant ground truth.

ArmTokens the model works fromAnswer quality (0–10 judge)Queries won or tied
Full local replayfull corpus (~1.7M)6.59—
TemporalStore managed pack~1.3k8.306 / 6

The margin is largest exactly where the answer lives in older history — the case recency-truncated replay cannot see. So the managed pack sends a tiny fraction of the tokens and answers better. Retrieval hit@k is 0.995–1.000 on LOCOMO and LongMemEval_s (7B).

Serving & scale

Fast at request time, correct under eviction, quick to recover.

Context serving is bounded temporal reads over a memory-first, multi-layer store — so latency stays low as the corpus grows, and cold state is promoted back into memory on read.

What we measureResult
ContextPack retrieval, p95 (1.2k-token budget)~17 ms
Retrieval hit-rate under an adaptive 1.2k→4k budget86% → 96%
Multi-user eviction + cold-read promotion correctness256k reads, 0 mismatches (81 users, 187k writes)
Warm reload after restart~168 ms

The eviction harness deliberately shrinks the memory budget to force memory→SSD→shared-storage eviction, then reads cold to force promotion — every value verified against ground truth, so a green run is also a correctness proof.

Read more

Full reports and write-ups.

ReportMemory benchmark write-upLOCOMO + LongMemEval, method and numbers. Parityvs a leading memory library, in detailHead-to-head parity analysis. Open sourcetemporalstore.ai benchmarksThe full OSS report and token-quota study.