LoCoMo v1, all 1,986 questions, Top 10Season 1 · Rank 3 of 7
Mem0 open source
Mem0 open source (in-process, full multi-signal setup)mem0/oss
Report a problem with this result (see the corrections policy)
The judge ruled the retrieved context correct for 1,579 of 1,986 questions (79.5%, 95% CI 77.1 to 81.7), using 459 tokens per question on average.
Figures
Context
- Tokens per question, mean
- 459
- Tokens, 95th percentile
- 543
- Accuracy per 1,000 tokens
- 173.2
- Contexts cut to the budget
- 0.0%
Speed
- Retrieval, median
- 592 ms
- Retrieval, 95th percentile
- 897 ms
- Ingestion, total
- 7 h 29 min (all conversations added up)
- Ingestion, per turn
- 4.59 s
Judge cost
- Top 10
- $10.24
- Top 30
- $15.36
- Whole run
- $25.60from price table 2026-10-04
Checks
- Abstention accuracy
- 94.2%
- No verdict (call failed)
- 0
- Flagged by the judge
- 25
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 | 420 of 446 | 94.2% | 91.4 to 97.0 | 446 | 0 |
| multi-hop | 131 of 282 | 46.5% | 41.4 to 52.0 | 456 | 0 |
| open-domain | 45 of 96 | 46.9% | 29.5 to 64.1 | 453 | 0 |
| single-hop | 722 of 841 | 85.9% | 83.3 to 89.3 | 461 | 0 |
| temporal | 261 of 321 | 81.3% | 75.3 to 86.6 | 478 | 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
- 20261101T010624Z-79759b
- Mode
- hosted runner
- Finished
- 1 Nov 2026
- Duration
- 2 h 35 min
- memjudge version
- 0.1.0
- Commit
- 896c4f7b13f7
- Image
- sha256:8671c6e11fb4468e182e776556e616e221df73a4f77051716772e8cd4238222e
- 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
- mem0/oss
- Written by
- MemJudge maintainers
- SDK
- mem0ai[nlp]==2.2.1
- Code hash
- sha256:2bea0837a48df5082a31b032cf620319ce603c7ea13f26727ca3e455980112fd
- Models
- gpt-6-luna, text-embedding-3-small
- llm model
- gpt-6-luna
- llm is reasoning model
- on
- embedder model
- text-embedding-3-small
- embedding dims
- 1536
Raw files
| File | SHA-256 |
|---|---|
| conversations.jsonl | b56d353cb45209bb866c411f79fba4db8c108d93281c92c323ab45c2a680e8f8 |
| questions-top10.jsonl | be7695c9f9557bf898e7a707c81387234763daaf61e2fea5a6f0b7c2d92cbe61 |
| questions-top30.jsonl | cd1e2c795f0278155f22e9cd8f073dea2623a7cdf997d2a0a55654ed59a3839a |
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 20261101T010624Z-79759b.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.