Our take
Relace is a small, specialised inference provider with a standout proprietary model built for extreme speed and a zero-retention data option. Its catalogue is narrow, and compliance details are largely undisclosed.
Use this for speed-sensitive proprietary inference where Relace Apply 3's throughput is the decisive factor. Pick it when zero-retention is a hard requirement and you do not need broad model choice. Skip it if you need verified SOC 2, a known headquarters location, or a deep catalogue.
- Extreme throughput on its proprietary model — Relace Apply 3 processes at 8,949 tps, far ahead of any other offer we track.
- Zero-retention data option available, plus a commitment not to train on prompts.
- On DeepSeek V4 Flash, the cheaper variant is also the faster one, making the best choice unambiguous.
- Compliance and geographic posture largely undisclosed: headquarters, EU endpoint and SOC 2 are all unverified in our data.
- Small catalogue — 10 tracked offers versus providers with 60-plus.
- Inconsistent price-performance on some model variants: Z AI GLM 5.3 Flash has three price points with no clear speed correlation, so the fastest is not the most expensive.
Point your tools here
We have not recorded what a router needs for Relace yet — no base URL and no docs link on file. That is our gap, not a sign Relace has no API; its own documentation is the place to look until we close it.
Privacy & data handling
These answers cover requests sent to Relace through OpenRouter, as OpenRouter records them, last read 58 min ago. For Relace's own API we hold no answer: check its terms. SOC 2 and the data processing agreement below are hand-checked, not part of that sync.
Prompt retention: none.