@1mbrain/benchmarks 0.1.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (69) hide show
  1. package/README.md +85 -0
  2. package/fixtures/1mbrain-focused-mini/1mbrain-focused-mini.json +928 -0
  3. package/fixtures/1mbrain-focused-mini/README.md +45 -0
  4. package/fixtures/adversarial-memory/dataset_claude_adversarial.json +3333 -0
  5. package/fixtures/adversarial-memory/dataset_gemini_adversarial_memory.json +2984 -0
  6. package/fixtures/balanced-mini/dataset_claude_balanced_mini.json +2077 -0
  7. package/fixtures/balanced-mini/dataset_gemini_balanced_mini.json +1995 -0
  8. package/fixtures/generate_datasets.js +1741 -0
  9. package/fixtures/graph-stress-hard/README.md +43 -0
  10. package/fixtures/graph-stress-hard/dataset_graph_stress_hard.json +4374 -0
  11. package/fixtures/graph-stress-hard/generate_graph_stress_hard.js +526 -0
  12. package/fixtures/realistic-medium/dataset_claude_realistic_medium.json +7462 -0
  13. package/fixtures/realistic-medium/dataset_gemini_realistic_medium.json +7277 -0
  14. package/fixtures/realistic-medium/gen_claude_medium.js +600 -0
  15. package/package.json +22 -0
  16. package/reports/benchmark_report.md +48 -0
  17. package/reports/benchmark_report_claude_adversarial.md +42 -0
  18. package/reports/benchmark_report_claude_adversarial_adaptive.md +42 -0
  19. package/reports/benchmark_report_claude_adversarial_adaptive2_fast.md +42 -0
  20. package/reports/benchmark_report_claude_adversarial_adaptive_fast.md +42 -0
  21. package/reports/benchmark_report_claude_adversarial_rerank.md +42 -0
  22. package/reports/benchmark_report_claude_balanced_mini.md +42 -0
  23. package/reports/benchmark_report_claude_balanced_mini_adaptive.md +42 -0
  24. package/reports/benchmark_report_claude_balanced_mini_adaptive2_fast.md +42 -0
  25. package/reports/benchmark_report_claude_balanced_mini_adaptive_fast.md +42 -0
  26. package/reports/benchmark_report_claude_balanced_mini_rerank.md +42 -0
  27. package/reports/benchmark_report_claude_realistic_medium.md +42 -0
  28. package/reports/benchmark_report_claude_realistic_medium_adaptive.md +42 -0
  29. package/reports/benchmark_report_claude_realistic_medium_adaptive2_fast.md +42 -0
  30. package/reports/benchmark_report_claude_realistic_medium_adaptive_fast.md +42 -0
  31. package/reports/benchmark_report_claude_realistic_medium_evidence_rerank_local.md +42 -0
  32. package/reports/benchmark_report_claude_realistic_medium_openai_evidence_rerank.md +41 -0
  33. package/reports/benchmark_report_claude_realistic_medium_openai_multi_signal.md +41 -0
  34. package/reports/benchmark_report_claude_realistic_medium_openai_multi_signal_scoped.md +41 -0
  35. package/reports/benchmark_report_claude_realistic_medium_openai_phase8_no_judge.md +42 -0
  36. package/reports/benchmark_report_claude_realistic_medium_openai_rankingpolicy.md +41 -0
  37. package/reports/benchmark_report_claude_realistic_medium_openai_stale_filter.md +41 -0
  38. package/reports/benchmark_report_claude_realistic_medium_openai_stale_filter_absence_fix.md +41 -0
  39. package/reports/benchmark_report_claude_realistic_medium_openai_write_time_invalidation.md +41 -0
  40. package/reports/benchmark_report_claude_realistic_medium_rerank.md +42 -0
  41. package/reports/benchmark_report_claude_realistic_medium_stale_filter_local.md +42 -0
  42. package/reports/benchmark_report_graph_stress_hard.md +42 -0
  43. package/reports/benchmark_report_graph_stress_hard_absence_fix.md +42 -0
  44. package/reports/benchmark_report_graph_stress_hard_adaptive.md +42 -0
  45. package/reports/benchmark_report_graph_stress_hard_evidence_rerank.md +42 -0
  46. package/reports/benchmark_report_graph_stress_hard_multi_signal_current_guardrail.md +42 -0
  47. package/reports/benchmark_report_graph_stress_hard_multi_signal_guardrail_fixed.md +42 -0
  48. package/reports/benchmark_report_graph_stress_hard_multi_signal_local.md +42 -0
  49. package/reports/benchmark_report_graph_stress_hard_multi_signal_scoped_guardrail.md +42 -0
  50. package/reports/benchmark_report_graph_stress_hard_multi_signal_vector_pure_guardrail.md +42 -0
  51. package/reports/benchmark_report_graph_stress_hard_phase8_sdk_guardrail.md +42 -0
  52. package/reports/benchmark_report_graph_stress_hard_rerank.md +42 -0
  53. package/reports/benchmark_report_graph_stress_hard_stale_filter.md +42 -0
  54. package/reports/benchmark_report_graph_stress_hard_write_time_invalidation.md +42 -0
  55. package/results/.gitignore +2 -0
  56. package/src/adapters/1mbrain.ts +317 -0
  57. package/src/adapters/keyword-embedding.ts +48 -0
  58. package/src/adapters/mem0.ts +124 -0
  59. package/src/adapters/qdrant.ts +214 -0
  60. package/src/adapters/unavailable.ts +49 -0
  61. package/src/adapters/vector-baseline.ts +149 -0
  62. package/src/datasets/focused-mini.ts +158 -0
  63. package/src/datasets/synthetic-agent-memory.ts +532 -0
  64. package/src/llm-evaluator.ts +262 -0
  65. package/src/metrics.ts +482 -0
  66. package/src/provider.ts +151 -0
  67. package/src/runner.ts +635 -0
  68. package/tsconfig.json +10 -0
  69. package/tsconfig.tsbuildinfo +1 -0

There are too many changes on this page to be displayed.

The amount of changes on this page would crash your brower.

You can still verify the content by downloading the package file manually.