Agentra LabsAgentra Labs DocsPublic Documentation

AgenticPlanning

Benchmarks

Performance characteristics of AgenticPlanning measured on a typical development machine (Apple Silicon, 16 GB RAM, single-threaded).

Performance characteristics of AgenticPlanning measured on a typical development machine (Apple Silicon, 16 GB RAM, single-threaded).

Operation Latency

OperationTypical LatencyNotes
Goal create<1 msIncludes index update
Goal get<1 msDirect HashMap lookup
Goal update (status change)<1 msIndex update is incremental
Decision create<1 ms
Decision crystallize<1 msShadow path serialization included
Commitment create<1 ms
Commitment fulfill/break<1 ms
Progress record<1 msIncludes momentum recalculation
Blocker scan1-3 msScales with active goal count
Dream generation2-5 msDepends on goal complexity
Singularity collapse5-15 msO(n) over active goals
File save5-20 msBTreeMap sorting + BLAKE3 + fsync
File load5-30 msIncludes full index rebuild

All CRUD operations (create, read, update, delete) on individual entities complete in sub-millisecond time. The engine's HashMap-based stores provide O(1) lookups.

Memory Footprint

ComponentSizeNotes
Per goal2-4 KBIncludes physics, feelings, progress history, tags, relationships
Per decision1-3 KBIncludes shadow paths, causal chain references
Per commitment1-2 KBIncludes entanglement references
Per dream1-2 KBScenarios, insights, seeds
PlanIndexes (base)~10 KB24 index structures, empty
PlanIndexes (100 goals)~50 KBGrows with entity count

Project-scale estimates:

ScaleGoalsDecisionsCommitmentsMemory
Small1055~80 KB
Medium503020~350 KB
Large1006040~500 KB
Very large500200100~2.5 MB

File Size Scaling

ContentApproximate Size
Empty engine~1 KB
5 goals~15 KB
10 goals, 5 decisions~25 KB
20 goals, 15 decisions, 10 commitments~60 KB
50 goals, 40 decisions, 30 commitments~150 KB
100 goals, 80 decisions, 60 commitments~300 KB

Growth is roughly linear:

  • Each goal adds ~2 KB to the file
  • Each decision adds ~1.5 KB
  • Each commitment adds ~1 KB
  • Index overhead grows proportionally to entity count

Index Performance

Index-assisted queries provide fast lookups regardless of total entity count:

QueryComplexityIndex Used
Goals by statusO(1)goals_by_status
Goals by priorityO(1)goals_by_priority
Goals by deadline (range)O(log n)goals_by_deadline (BTreeMap)
Root goalsO(1)root_goals
Active/blocked goalsO(1)active_goals / blocked_goals
Decisions by goalO(1)decisions_by_goal
Commitments due soonO(log n)commitments_by_due (BTreeMap)
Urgent itemsO(1)urgent_items

Scaling Guidance

  • Sweet spot: 5-100 goals. All operations are sub-millisecond except singularity collapse.
  • Comfortable range: Up to 1,000 entities total. File sizes stay under 3 MB. Load times under 100 ms.
  • Upper bound: ~10,000 entities. Linear scaling holds. File operations become the bottleneck (~200+ ms saves).
  • Not designed for: Millions of entities. Use a database-backed planning system for that scale.

The engine is single-threaded (PlanningEngine is !Sync). All benchmarks reflect sequential operation. MCP server handles one tool call at a time.

Checksum Overhead

BLAKE3 hashing adds minimal overhead to save operations:

Entity CountChecksum Time% of Total Save
10<1 ms~5%
100~1 ms~10%
1000~3 ms~15%

BTreeMap sorting (for deterministic checksums) is the larger contributor to save latency at scale.