Ellipsis

https://www.ellipsis.dev

Ellipsis is a cloud platform that runs managed coding agents (powered by Claude) on behalf of engineering teams. Each agent executes inside its own isolated Linux sandbox in Ellipsis's cloud, with credentials scoped below any developer. Agents are defined as YAML files in the user's repository and triggered on schedules, PR events, mentions, or via API, and are used for code review, automated fixes, dependency bumps, and similar engineering tasks.

Pricing model Usage-based; personal accounts pay no sandbox compute, organizations pay for allocated CPU and memory while running plus model charges; Ellipsis applies a 10% platform fee to credit purchases
Pricing The headline compute price as the provider states it. Sorting ranks prices stated per vCPU over time, converted to hourly. Other pricing models sort after, unranked. $0.142 / vCPU-hour (Ellipsis Cloud tier)
Free while idle Whether a stopped, paused, or sleeping sandbox costs nothing. ✓
Elastic Resources flex with what the sandbox actually uses, and the bill follows: yes when the provider bills on active or observed use, no when capacity is reserved or allocated and billed while it runs. ✕
Memory snapshots Whether RAM state survives a pause and resume, not just disk. ✕
Wake on request Whether a stopped sandbox wakes automatically on inbound traffic. –
Start / resume Typical time from create or resume to running. Sorting ranks stated times, converted to seconds; claims with no figure sort after, unranked. ~8 s (warm snapshot boot); first session image build ~3m40s
Max runtime The longest a sandbox may run. 24 h
Isolation The isolation technology between sandboxes. Isolated Linux VM/machine; exact underlying technology is not publicly specified
GPUs Whether GPU instances are available for sandboxes. –
Docker Whether Docker containers can run inside the sandbox. –
SDKs –

✓ yes · ✕ no · – no cited public fact.

Ellipsis states the basis for its 10% platform fee three ways: credit purchases before discounts in the Pricing docs, token cost in the pricing page's comparison table, and a session's token and compute cost in that page's FAQ. The Pricing docs charge organizations for allocated CPU and memory while a sandbox runs, including while it waits for the next message; the pricing page says nothing runs idle, so nothing bills idle.

Sources

Last verified 2026-10-06. Corrections land through the update log.