Everstack
everstack.ai · Compute · 1 record · JSON
Everstack Compute
trialemailno cardphoneDvolume-quota
| Free tier | Sandbox compute bills per second at USD 0.0333 per hour for Nano (0.5 vCPU, 0.5 GiB), USD 0.0666 for Small (1 vCPU, 1 GiB), USD 0.1332 for Medium (2 vCPU, 2 GiB), USD 0.2664 for Large (4 vCPU, 4 GiB) and USD 0.5328 for XL (8 vCPU, 8 GiB). 'Every new organization receives a one-time $5 sandbox compute credit with no payment method required.' The platform plan is USD 20 a month for Build with 3 seats, USD 100 a month for Scale with 10 seats. unit minutes; metered on wall-clock; one-time-credit |
|---|---|
| Price | none stated |
| Per-object cap | not applicable |
| Retention | not applicable |
| Egress | not stated |
| Identity | email no payment method required: the pricing page states "Every new organization receives a one-time $5 sandbox compute credit with no payment method required", so there is no card and no hold; the credit is granted per organisation. The record does not state what signup asks for — an email address, a phone number or a company name — nor whether debit or prepaid cards are accepted later, which does not arise while no payment method is taken |
| Limits |
Memory is 1 GiB per vCPU on every size, so a workload that needs 4 GiB of RAM has to buy a 4 vCPU box at USD 0.2664 per hour; the USD 5 credit is one per organization and does not renew. |
| Binding | volume-quota Memory is 1 GiB per vCPU on every size, so a workload that needs 4 GiB of RAM has to buy a 4 vCPU box at USD 0.2664 per hour; the USD 5 credit is one per organization and does not renew. |
| Interface | rest inferred from the tooling the provider documents; not probed |
| Tooling | not stated |
| Composes with | The sandbox sizes line up with the whole platform: the same account supplies an AI gateway over 200 models, a vector memory store and the sandbox, so an agent can be built without a second vendor, and the sandbox is the only part billed by the second. |
| Reliability | no reliability event recorded |
| Notes | Rate per vCPU-hour works out to USD 0.0666 flat, which is a quarter above ix's USD 0.0288 and double Bunny's USD 0.02, and the RAM ratio is the reason: 1 GiB per vCPU is tight for a build that compiles anything. The USD 5 credit at USD 0.0666 per hour is 75 hours of Small. Whether sandbox compute requires the USD 20 Build plan is not stated on the pricing page; the credit is offered to a new organization with no payment method, which implies at least a trial path without one. Isolation technology not read. Documentation exposes an llms.txt and an OpenAPI index, so the API surface is testable without a browser. |
| Evidence rung | documented |
| Checked | 2026-10-22 (live); stale on 2026-11-21 |
| Parsed numbers |
|
Evidence (4)
| date | source | what it proved |
|---|---|---|
| 2026-10-22 | everstack.ai/pricing | sandbox compute per second: Nano USD 0.0333 per hour at 0.5 vCPU and 0.5 GiB, Small USD 0.0666 at 1/1, Medium USD 0.1332 at 2/2, Large USD 0.2664 at 4/4, XL USD 0.5328 at 8/8 |
| 2026-10-22 | everstack.ai/pricing | 'Every new organization receives a one-time $5 sandbox compute credit with no payment method required' |
| 2026-10-22 | everstack.ai/pricing | Build USD 20 per month with 3 seats, Scale USD 100 per month with 10 seats, hosted browser sessions USD 0.05 per browser hour with a 1 minute minimum |
| 2026-10-22 | docs.everstack.ai/getting-started/configuration/ | the sandbox backend is selectable between docker and kubernetes, with defaults of max_cpu 4.0 and max_memory_mb 4096 |
More fields (1)
{
"wave2": {
"class": "platform-with-sandbox",
"free_tier_kind": "one-time-credit",
"shape": {
"vcpu": 2,
"ram_gib": 2,
"disk_gib": null,
"gpu_class": null,
"gpu_count": null
}
}
}Unknown or withheld values (8)
$.egress.meteredegress-unknown$.egress.termsnot-stated$.per_object_cap.notesprocess$.per_object_cap.statednot-applicable$.reliability_notenone-observed$.retention.notesprocess$.retention.statednot-applicable$.toolingnot-stated