| What it does | Agent memory that learns from experience
| Composable computers for devs and agents
|
|---|
| Stage | Startup
| Not given
|
|---|
| Founded | 2024
| Not given
|
|---|
| Based in | US
| NL
|
|---|
| Pricing model | Freemium
| Not given
|
|---|
| Pricing | 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
| Credit-based pricing: €0.039 per vCPU-hour (active only), €0.012 per GiB-hour for RAM (memory in use), €0.00008 per GiB-hour for disk (written space only). New accounts receive €30 in free credits. A typical 2 vCPU / 8 GiB machine costs approximately €0.03 per hour while running (with 10% vCPU activ
|
|---|
| Key features | - 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
| - Hardware-isolated Linux VMs with persistent state between sessions
- Snapshots and checkpoints for machine state management
- Automatic hibernation after 4 hours idle, paying for disk only
- Pre-installed tools: Claude Code, Docker, Node, Python, Go, git, Chrome, ffmpeg, nginx
- Public HTTPS URL per machine with custom domain support
- Egress allowlist for network security and isolation
- CLI, API, Python and TypeScript SDKs for automation
- Integration with GitHub, Linear and Slack across all machines
|
|---|
| What makes it different | - 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
| - Hardware-isolated real VMs instead of containers, with full kernel isolation
- Persistent machines remember running processes and state between disconnects
- Pay only for active resources used, not provisioned size
- Free self-hosting up to 16 vCPUs on your own infrastructure
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 68 ▲ +1 (up) since last check | 44 ● no change since last check |
|---|
| Trust Flow (Majestic) | 18 | 8 |
|---|
| Citation Flow (Majestic) | 47 | 37 |
|---|
| Referring domains (Majestic) | 1378 | 151 |
|---|
Vectorize
boxd