LoCoMo v1, all 1,986 questions, Top 10Season 1 · Rank 5 of 7
BM25 keyword retrieval over raw turns
BM25 keyword retrieval over raw turns (baseline)baselines/bm25
Report a problem with this result (see the corrections policy)
The judge ruled the retrieved context correct for 1,275 of 1,986 questions (64.2%, 95% CI 62.9 to 65.5), using 451 tokens per question on average.
Figures
Context
- Tokens per question, mean
- 451
- Tokens, 95th percentile
- 586
- Accuracy per 1,000 tokens
- 142.3
- Contexts cut to the budget
- 0.0%
Speed
- Retrieval, median
- 3.21 ms
- Retrieval, 95th percentile
- 5.96 ms
- Ingestion, total
- 262 ms (all conversations added up)
- Ingestion, per turn
- 45 µs
Judge cost
- Top 10
- $10.17
- Top 30
- $15.29
- Whole run
- $25.46from price table 2026-10-04
Checks
- Abstention accuracy
- 95.5%
- No verdict (call failed)
- 0
- Flagged by the judge
- 20
- Evidence recall
- 59.1%
Where it fails
Accuracy by question category, with 95% confidence intervals. A low category is where the memory system misses the context the question needs.
| Category | Correct | Accuracy | 95% CI | Tokens | No verdict |
|---|---|---|---|---|---|
| adversarial | 426 of 446 | 95.5% | 94.0 to 97.2 | 453 | 0 |
| multi-hop | 55 of 282 | 19.5% | 14.7 to 23.9 | 452 | 0 |
| open-domain | 30 of 96 | 31.3% | 19.0 to 43.2 | 424 | 0 |
| single-hop | 556 of 841 | 66.1% | 63.0 to 69.8 | 454 | 0 |
| temporal | 208 of 321 | 64.8% | 58.0 to 71.7 | 448 | 0 |
Every question
One square per question, grouped by category. Select a square to read the question, the gold answer, the judge's reason and the context it was given.
Loading questions…
Retrieval latency
The share of questions whose context came back within a given time. Further left is faster.
Judge
A season is one fixed set of judging rules: one pinned judge model, its settings and its prompt rule on every question. When the rules change a new season starts, and earlier runs are judged again from their stored contexts, without re-running the memory system. Seasons
- Model
- claude-sonnet-5-5
- Route
- anthropic
- Season
- Season 1
- Effort
- medium
- Output limit
- 4,000 tokens
- Prompt
- v3 sha256:ce6d9750331c79d1bb3a0737109c458fef4ba9fd8cf929c75c1c1cf9407a4b78
- Served by
- claude-sonnet-5-5
How this result was produced
Everything below is part of the signed result. Change any of it and the signature no longer matches.
Run
- Run id
- 20261101T003952Z-a4ec8e
- Mode
- hosted runner
- Finished
- 1 Nov 2026
- Duration
- 18 min 03 s
- memjudge version
- 0.1.0
- Commit
- 4fabdb2909e6
- Image
- sha256:2e89150c73f10947df66c017dd905ae0208b9a839d87f26a940f441bccdbeb9f
- Context budget
- 8,192 tokens (o200k_base)
- Conversations at a time
- 4
- Submitter affiliated
- no
Dataset
- Name
- locomo v1
- Questions
- 1,986
- Licence
- CC-BY-NC-4.0
- Corrections applied
- 0
- Hash
- sha256:9c7812f2f45ce12b06cd9d3fc55736c3a8a83d2b263c11f28c292c475f6ddf38
- Source file hash
- 79fa87e90f04081343b8c8debecb80a9a6842b76a7aa537dc9fdf651ea698ff4
Connector
- Connector
- baselines/bm25
- Written by
- MemJudge maintainers
- SDK
- none (plain HTTP)
- Code hash
- sha256:1ef7f15e03d56b16b7c8535b4862e2ac398cf13925bee541a221e44179597b0d
- k1
- 1.5
- b
- 0.75
Raw files
| File | SHA-256 |
|---|---|
| conversations.jsonl | 355db4834e174b0b5630afea7a818e966d4403453ba9e5f74dea9568410c8973 |
| questions-top10.jsonl | d34d5d2cf1fd0e19cca5f9087bd46d9782a047c52f73dd05791f2311e748e889 |
| questions-top30.jsonl | 6dd8f5027eda5262ea62e1d19ed756e2cdead2428ef9ab24228b1806fe0f9818 |
Check it yourself
The signed result is signed with key 478df5eb628bfa74, one of the trusted keys published in the leaderboard repository. Download it and verify it with the memjudge command-line tool; add --artifactswith the raw files to check those too. How signing works.
memjudge verify 20261101T003952Z-a4ec8e.jsonQuestions, answers and conversation excerpts on this page come from LoCoMo (Maharana et al., 2024, Snap Research) and are used under CC BY-NC 4.0 (non-commercial use only). They are reformatted, and corrected where MemJudge's errata apply. MemJudge's own verdicts and scores are CC BY 4.0.