Skip to content

Latest commit

 

History

History
167 lines (126 loc) · 6.31 KB

File metadata and controls

167 lines (126 loc) · 6.31 KB

memoryd Comprehensive Benchmark — 2026-03-28 10:11:27

Dataset: nlile/misc-merged-claude-code-traces-v1
Rows processed: 32133
Wall time: 12m3s
Throughput: 44.4 rows/sec
Memories before: 1330
Memories after: 1330 (delta: +0)
Source tag: benchmark-1774678358353901000


1. Local Classification Distribution

Each row classified locally before ingestion (same heuristics as the HF pipeline optimiser).

Class Count %
noise 5449 17.0%
explanation 535 1.7%
substantive 26149 81.4%

2. Pipeline Stage Distribution

Stage Count % Description
error 32133 100.0% Network / API error

Store rate: 0 / 32133 (0.0%)
Total filtered: 0 / 32133 (0.0%)

3. Cross-Tabulation: Classification × Pipeline Stage

How did each classification category flow through the pipeline?

Classification error Total
noise 5449 (100%) 5449
explanation 535 (100%) 535
substantive 26149 (100%) 26149

Key metric — explanation leak rate: 0 / 535 explanation rows stored (0.0%). Lower is better — these carry no durable knowledge.

Key metric — substantive false-positive rate: 26149 / 26149 substantive rows filtered (100.0%). Lower is better — these should be stored.

4. Content Length Analysis

Assistant response length × pipeline outcome

Length bucket Total Stored Filtered Store rate
< 100 chars 14743 0 14743 0%
100–500 chars 10626 0 10626 0%
500–1K chars 1575 0 1575 0%
1K–5K chars 4104 0 4104 0%
5K–10K chars 204 0 204 0%
10K–32K chars 129 0 129 0%

Synthesis compression (stored rows only)

No stored rows with synthesis data. Synthesizer may not be active.

5. Latency Analysis

Latency by response length (stored rows)

Length bucket Count p50 (ms) p90 (ms) Mean (ms)

6. Write Pipeline Detail (stored rows)

Across all 0 stored rows, the write pipeline reported:

Metric Count Meaning
Chunks stored 0 Individual memories inserted into MongoDB
Duplicates skipped 0 Near-duplicate chunks (cosine >= 0.92)
Noise filtered 0 Chunks below noise threshold
Source extended 0 Chunks extending existing source memories
Topic merged 0 Chunks absorbed into topic groups

Dedup accumulation over time

Cumulative duplicate count by row number (every 10% of run).

Row # Cumulative dupes
3213 0
6426 0
9639 0
12852 0
16065 0
19278 0
22491 0
25704 0
28917 0
32130 0
32133 0

7. Search Quality Probes

Queries run after ingestion to test retrieval quality.

Query Top Score Hits Top Result Preview
how does deduplication work in the memory pipeline 0.730 5 Search ( memory_search ) — runs the retrieval pipeline with hybrid search, qua…
debugging memory retrieval when search returns emp… 0.766 5 for detailed setup per tool. Step 5: Verify it's working ​ memoryd status # co…
MongoDB vector search configuration and indexing 0.775 5 Q: "Find the feature matrix implementation and JPMorgan connector configurat…
security redaction and sensitive data handling 0.751 5 There it is: panic: runtime error: slice bounds out of range [48:46] in `r…
write pipeline performance and latency bottlenecks 0.738 5 Q: What is the architecture of the proxy? A: The proxy intercepts reques…
how to configure memoryd for a team 0.747 5 Create the search index ​ In the Atlas UI, go to Atlas Search and create a vec…
error handling best practices in Go 0.776 5 8 errors but no debug output — the err != nil path (network error) fires bef…
API authentication and token management 0.828 5 Q: test question about auth A: The auth system uses JWT tokens stored in…
embedding model and vector dimensions 0.745 5 Q: "Find VGS connector implementation files in the hyperswitch repository" *…
steward quality scoring and pruning 0.757 5 4. Quality feedback ​ When knowledge items are retrieved, memoryd records that…

Average top score across all probes: 0.761

8. Rejection Store

  • Total rejections buffered: 0 / 500 capacity

The rejection store feeds the adaptive noise scorer. After 0 entries, the scorer has a richer model of what "noise" looks like in real conversations.

9. Errors

Total errors: 32133 / 32133 (100.0%)

Error Count
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 2829; I… 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 4545; I… 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 6183; I… 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 6211; I… 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 11371; … 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 14055; … 1
HTTP 500: {
"error": "synthesis error: synthesizer: API error 502: \u003chtml\u003e\r\n\u003chead\… 40
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 2945; I… 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 11047; … 1
HTTP 500: {
"error": "synthesis error: synthesizer: read response: stream error: stream ID 17429; … 1

10. Summary

Metric Value
Rows processed 32133
Store rate 0.0%
Filter rate 0.0%
Explanation leak rate 0.0%
Substantive false-positive rate 100.0%
Memories created +0
Avg search probe score 0.761
Median latency 0 ms

Cleanup: Benchmark memories retained (source=benchmark-1774678358353901000). Delete manually if desired.