Engine Pricing
Context compression: We intelligently abstract and compress input context, significantly reducing effective token usage. Code and research data is processed fully for precision on Palo Research and Palo Research engines.
Subscription Plans: Architect ($20/mo) includes 120M tokens, 10GB of storage, weekly usage visibility, and 1 free migration + 1 free memory merge. SuperBloom ($250/mo) includes 1B tokens, 10TB of storage, 50 Private Memory Space slots, the most detailed weekly usage visibility, unlimited migrations + memory merging, and per-user_id analytics. Overage billed at rates below.
| Engine / API Name | Mini (Palo Nano) | Palo Bloom | Palo Research | Palo Research |
|---|---|---|---|---|
| Blended I/O Rate per 1M tokens | $0.30 | $0.90 | $2.10 | $2.90 |
| Memory Traversal optional, per 1M tokens | $0.18 | $0.54 | $1.26 | $1.76 |
| Memory Mapping optional, per 1M tokens | $0.24 | $0.72 | $1.68 | $2.25 |
Monthly Cost Calculator
Estimated Monthly Total
$0.00