Our take
Inception is a small, privacy-focused inference provider that hosts only its own Mercury model family in two sharply separated price and speed tiers. It offers a zero-retention option and does not train on prompts, but its compliance credentials and geographic footprint are unverified in our data.
Use this when zero-retention is a hard requirement, or for cost-sensitive inference where the Mercury 2.5 Preview tier is fast enough. Pick it if you accept a single-model-family catalogue and variable throughput. Skip it if you need verified SOC 2, a confirmed EU endpoint, or a choice of model families.
- Strong privacy posture: zero-retention is available and prompts are not used for training.
- Mercury 2.5 Preview is dramatically cheaper than Mercury 2 within the same family.
- Mercury 2.5 Preview showed more than double the throughput in one run versus another, suggesting tiered or variable infrastructure.
- Extremely narrow catalogue: only 3 tracked offers, all from one model family.
- Compliance and geographic footprint are undocumented: SOC 2, EU endpoint and headquarters country are all unverified in our data.
- Mercury 2 costs several times more than its preview variant at the same measured throughput.
Point your tools here
We have not recorded what a router needs for Inception yet — no base URL and no docs link on file. That is our gap, not a sign Inception has no API; its own documentation is the place to look until we close it.
Privacy & data handling
These answers cover requests sent to Inception through OpenRouter, as OpenRouter records them, last read 4 hours ago. For Inception'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.