- Free allowance
- No free GPU tier is published; list rates per GPU-hour from the GPU catalog read on 2026-11-03: RTX 3090 USD 0.18, RTX 4090 USD 0.35, RTX 5090 USD 0.50, A100 80GB USD 1.10, RTX Pro 6000 USD 1.55, H100 SXM USD 2.60, B200 USD 6.30, AMD Instinct MI355X USD 6.30, B300 USD 8.49.
- Verdict
- paidno free allowance; listed for its price
- Cost
- no price stated
- Identity
- cardcard required, no phone
- Binding limit
- noneno named constraint binds
- Evidence
- documented6 sources; a first-party page was read and the sentence quoted
- Freshness
- stale on 2026-12-04checked 2026-11-03 (live); stale on 2026-12-04
| Free tier | no free GPU tier is published; list rates per GPU-hour from the GPU catalog read on 2026-11-03: RTX 3090 USD 0.18, RTX 4090 USD 0.35, RTX 5090 USD 0.50, A100 80GB USD 1.10, RTX Pro 6000 USD 1.55, H100 SXM USD 2.60, B200 USD 6.30, AMD Instinct MI355X USD 6.30, B300 USD 8.49. The catalog lists no L40S and no L4. 'No commitment, no quota request, no reservation.' unit none; metered on wall-clock; multi-GPU VMs multiply the per-GPU rate; the aggregator shows a community and a verified lane for some cards (L4 at USD 0.43 community, USD 0.55 verified) that the provider catalog page does not list |
|---|
| Price | none stated |
|---|
| Binding | none No quota is stated; the rate list is the constraint, and capacity is on distributed hardware that can go down. |
|---|
| Identity | card The catalog says 'Add a card and make a deposit to test.' The terms page says an account must be registered; no phone or ID document is stated on the pages read. The sign-up page itself answered 404 and was not read. |
|---|
| Egress | not published on the pages read |
|---|
| Interface | rest inferred from the index, which names a /v1 API and an OpenAI-compatible gateway; not probed |
|---|
| Tooling | GPU VMs and container pods from a web console; an OpenAI- and Anthropic-compatible inference gateway at inference.openrelay.inc/v1; the index says the service was formerly VectorLay and is in early access |
|---|
| Composes with | Per-GPU hourly rates for A100 80GB, H100 SXM and RTX 4090 with no commitment, next to an OpenAI-compatible inference API on the same account. |
|---|
| Reliability | The terms page states 'Our Services run on distributed consumer hardware, and individual nodes may experience downtime.' The index states early access. |
|---|
| Notes | Found via computeprices.com (provider slug openrelay). Not to be confused with Metered Open Relay (storage/metered-open-relay-at-the-advertised-20-gb), a different service. Not read: the L4 and L40S rows, which the aggregator lists but the provider catalog does not, and the sign-up and deposit minimum. |
|---|
| Parsed numbers | |
|---|