AutoGPT Memory
Community-VerifiedMemory subsystem within the AutoGPT autonomous agent framework. Provides file-backed and vector-store memory for persistent task context across agent execution cycles.
Scores from 0–100. Higher is better. LLM Baseline (no memory system) scores 57.6%. How we calculate this →
TrackAgent Memory
Track Index
64.8/100
Based on 5 benchmarks.
Benchmark Results
| Benchmark | Score | Status | Receipt |
|---|---|---|---|
| Knowledge Retrieval | 100.0 | Verified | View |
| Truth Arbitration | 80.0 | Verified | View |
| Memory Poisoning | 0.0 | Verified | View |
| Budget Curves | 100.0 | Verified | View |
| Reliability | 44.0 | Verified | View |
| Other Benchmarks | |||
| LongMemEval | Not applicable — outside Agent Memory track | ||
| LoCoMo | Not applicable — outside Agent Memory track | ||
| Knowledge Scale | Not applicable — outside Agent Memory track | ||
Relative Performance vs All Benchmarked Systems
vs 16 scored systemsEach dot is a system. Amber dot is AutoGPT Memory. Amber line = LLM Baseline (no memory).
Overall47.425th percentile
No memory: 57.6%gbrain
Recall73.125th percentile
No memory: 57.6%gbrain
Temporal37.438th percentile
No memory: 57.6%gbrain
Reasoning33.138th percentile
No memory: 57.6%gbrain
Bench'd Memory Index
The BMI combines accuracy (70%) and efficiency (30%) into a single production-weighted score. Formula is public and versioned.
38.2
/ 100
#4 of 8 systemsTop 50%
Accuracy (70%)47.4
Efficiency (30%)52.6
Efficiency Metrics
Avg Latency
Average time to retrieve memories and generate an answer. Lower is better.
Tokens / Correct
Average tokens consumed per correctly answered question. Lower means more efficient.
Recall Tokens
Average tokens returned by the memory system per query. Lower means tighter retrieval.
Per-Capability Score Matrix
| Dimension | Budget Curves | Knowledge Retrieval | LongMemEval | Memory Poisoning | Reliability | Truth Arbitration |
|---|---|---|---|---|---|---|
| Recall | -- | -- | 73.1 | -- | -- | -- |
| Temporal | -- | -- | 37.4 | -- | -- | -- |
| Reasoning | -- | -- | 33.1 | -- | -- | -- |
| Hallucination | -- | -- | -- | -- | 0.0 | -- |
| Stale Memory | -- | -- | -- | -- | 71.4 | -- |
| Entity Confusion | -- | -- | -- | -- | 100.0 | -- |
| Deletion | -- | -- | -- | -- | 0.0 | -- |
| Budget 1000 | 100.0 | -- | -- | -- | -- | -- |
| Budget 10000 | 100.0 | -- | -- | -- | -- | -- |
| Budget 2000 | 100.0 | -- | -- | -- | -- | -- |
| Budget 500 | 100.0 | -- | -- | -- | -- | -- |
| Budget 5000 | 100.0 | -- | -- | -- | -- | -- |
| Conflict resolution | -- | -- | -- | -- | -- | 80.0 |
| Document retrieval | -- | 100.0 | -- | -- | -- | -- |
| Injection resistance | -- | -- | -- | 0.0 | -- | -- |
| Knowledge update | -- | 100.0 | -- | -- | -- | -- |
| Multi page | -- | 100.0 | -- | -- | -- | -- |
| Semantic search | -- | 100.0 | -- | -- | -- | -- |
| Overall | 100.0 | 100.0 | 47.4 | 0.0 | 44.0 | 80.0 |
Per-Benchmark Breakdown
| Benchmark | Verified | Nuance |
|---|
Performance Over Time — LongMemEval
2026-05-11 to 2026-05-13AutoGPT47.4
Most often compared with
Add badge to your README
Show your Bench'd score on your GitHub repo.
Markdown
[](https://benchd.ai/system/autogpt-memory)
HTML
<a href="https://benchd.ai/system/autogpt-memory"><img src="https://img.shields.io/badge/Bench'd_BMI-38.2-D9982B?style=flat" alt="Bench'd Verified: 38.2 BMI" /></a>