Cloud agent sandbox with persistent filesystem and on-demand compute
Agent memory that learns from experience
Stage
MVP
Startup
Founded
Not given
2024
Based in
US
US
Pricing model
Paid
Freemium
Pricing
Compute: $0.015 per GiB-hour (approximately $0.000004167 per GiB-second, billed by the second). Storage: $0.02 per GiB-month for retained incremental RootFS data and memory checkpoints. Network ingress and egress currently free. Object-store GET and PUT requests free. New users receive $5 free credi
Self-hosted: Free, MIT licensed. Hindsight Cloud: Pay-as-you-go with free credits to start. Retain (store memories) $10.00 per million tokens, Recall (retrieve memories) $0.75 per million tokens, Reflect (synthesize patterns) $0.05 per call, Iris Extract (structured extraction) $7.50 per million tok
Key features
Automatic filesystem growth as storage fills
Pause compute without losing work or paying compute charges
Resume to same environment on demand
Access files while paused without waking compute
Public HTTP services with auto-resume capability
Command-line interface (s0 CLI) for agent-native workflows
Incremental checkpoints to avoid full filesystem copies
Scoped API keys and team-based access controls
On-demand file reads without downloading entire filesystem
Per-user persistent memory with preferences, history and decisions kept separate
Cross-session persistence allows context to survive weeks between interactions
Fast parallel retrieval in under 100ms through four strategies running simultaneously
Works with any LLM without losing learned context when swapping models
Learns from mistakes and corrects behavior on repeated problems
Automatic pattern detection synthesizes individual facts into consolidated knowledge
Conflict detection preserves both old and new states with timestamps when facts change
Integrates with 20+ coding agents including Claude Code, Cursor, and Copilot
What makes it different
Persistent filesystem across pause/resume without duplicating data
Pay only for compute time used; paused sandboxes have zero compute charges
Automatic filesystem growth so agents can install and persist unlimited tools
Same API for both human users and AI agents
94.6% retrieval accuracy on LongMemEval-S, beating next-best system at 74.0%
MIT licensed open source with no usage limits, telemetry or restrictions
Reflection layer automatically builds judgment and mental models over time without manual tagging
Entity resolution handles hundreds of entities with 99.5% query reduction
Vectorize