Mem

Memory systems, readers, and writers.

3 studies
6 comparisons
189 original files

All memory comparisons

Complete + bound counts complete answers with source-matched quotes. Each row is a separate test.

Memory comparison overview
Comparison	Complete + bound	Resources	Evidence scale
Readers · fixed-context replay → Luna / DeepSeek Flash 1017	DeepSeek Flash 101711/13 Luna medium11/13	$0.004811–$0.0193694.54–5.05 s / request	13 inputs / profile2 profiles · one pass
Readers · fixed-context replay → Luna reasoning effort	Luna low10/13 Luna medium11/13 Luna none11/13	$0.004149–$0.0047273.63–4.95 s / request	13 inputs / profile3 profiles · one pass
Systems · native retrieval → Native memory systems	OpenViking11/13 RAGFlow rc111/13	$0.006922–$0.024261Time: withheld from comparison	13 inputs / profile2 profiles · one pass
Systems · nightly version check → Native memory systems	RAGFlow nightly11/13	Cost not reported4.74 s / request	13 inputs / profile1 profile · one pass
Writers · extraction replay → Luna / DeepSeek Flash 1017 Viewing	Quality not measuredDeepSeek Flash 1017 · Luna medium	$0.001142–$0.0029144.57–8.31 s / request	3 inputs / profile2 profiles · one pass
Writers · extraction replay → Luna reasoning effort	Quality not measuredLuna low · Luna medium · Luna none	$0.000709–$0.0011203.52–7.72 s / request	4 inputs / profile3 profiles · one pass

Separate tests · no combined ranking.

说明

Costs are workload totals in USD; resource ranges span profiles within that row. Quote binding does not establish semantic entailment.

Writers · extraction replay

Luna / DeepSeek Flash 1017 · Extraction prompt replay

↑ All comparisons

Refine comparison

Study

Test

3 inputs / profile · 2 profiles · October 20, 2026

Test method

Resource use for the same inputs. Semantic writer quality was not measured.

Results & evidence ↓

Mean time

Seconds / input · lower is better

DeepSeek Flash 10174.57 s
Luna medium8.31 s

Settled cost

USD / test · lower is better

DeepSeek Flash 1017$0.002914
Luna medium$0.001142

Results and evidence

Sort columns. Open Details for the profile and original record.

Expand a row for all metrics

Sort

Metric

Writer · 3 paired inputs. Values are specific to this test. Actual settled USD for the three paired writer calls. Parsing compatibility is distinct from native ingestion acceptance.
Profile version				Evidence	Details
Luna mediumgpt-6-luna	3	8.312 s	$0.001142	Result evidence Luna medium Writer · 3 paired inputs · 3 inputs per profile Profile Time s · mean Cost USD total DeepSeek Flash 1017 4.569 s $0.002914 Luna mediumSelected 8.312 s $0.001142 Resource use for saved extraction outputs. Writer quality is not scored. All recorded measures Inputs per profile 3 Time s · mean 8.312 s Cost USD total $0.001142100 Configuration & original record Recorded date 2026-10-20 Model openai/gpt-6-luna Provider OpenAI Effort medium OutputAllowance 128,000 tokens Control Retained RAGFlow nightly calls Paired extraction prompt replay Actual settled USD for the three paired writer calls. Parsing compatibility is distinct from native ingestion acceptance. One diagnostic pass with real provider calls. Sort order is descriptive. Source binding alone does not establish semantic support. Open original metrics.json ↗
DeepSeek Flash 1017deepseek-v4.1-flash-1017	3	4.569 s	$0.002914	Result evidence DeepSeek Flash 1017 Writer · 3 paired inputs · 3 inputs per profile Profile Time s · mean Cost USD total DeepSeek Flash 1017Selected 4.569 s $0.002914 Luna medium 8.312 s $0.001142 Resource use for saved extraction outputs. Writer quality is not scored. All recorded measures Inputs per profile 3 Time s · mean 4.569 s Cost USD total $0.002913504 Configuration & original record Recorded date 2026-10-20 Model deepseek/deepseek-v4.1-flash-1017 Provider DeepSeek; no fallback Effort low OutputAllowance 393,216 tokens Paired extraction prompt replay Actual settled USD for the three paired writer calls. Parsing compatibility is distinct from native ingestion acceptance. One diagnostic pass with real provider calls. Sort order is descriptive. Source binding alone does not establish semantic support. Open original metrics.json ↗

One pass · — Not measured.

说明

USD totals are rounded to 6 decimals. The value tooltip and original record retain the recorded precision. Sorting is descriptive.

Actual settled USD for the three paired writer calls. Parsing compatibility is distinct from native ingestion acceptance.

Recorded finding

Three paired writer inputs. No semantic writer-quality score was measured. The extra Chinese input is excluded from this table.

Protocol and metric definitions

Three identical English extraction requests. The extra Chinese DeepSeek sample is available as a case and is excluded from this paired table.

Synthetic inputs · real provider calls · one pass · no repeated-trial interval.

Inputs: paired extraction requests. Time: mean request duration. Cost: settled USD for the selected workload. Output counts and JSON parsing are not quality scores.

Full conditions and limitations

Results