Benchmark suite for NIMA memory system. External to nima-core โ testing only.
| Benchmark | Status | NIMA Score | Supermemory |
|---|---|---|---|
| LongMemEval | ๐ง Ready to run | TBD | 81.6% |
| LoCoMo | ๐ Planned | โ | #1 |
| ConvoMem | ๐ Planned | โ | #1 |
cd longmemeval
bash download_data.sh
python eval.py --limit 50