Inference.
The decentralized AI inference marketplace
An OpenAI-compatible API served by independent GPU nodes earning $QAIS on Arbitrum. Compute priced per token, settled on-chain — no centralized provider.
A holdings company building the decentralized-AI economy — and the products on top of it.
Intelligence is becoming infrastructure.
Infrastructure should not have a landlord.
The decentralized AI inference marketplace
An OpenAI-compatible API served by independent GPU nodes earning $QAIS on Arbitrum. Compute priced per token, settled on-chain — no centralized provider.
Meet Conquistar
Give Codex a bounded task. Check TypeScript, JavaScript, Python, Rust or Go changes against explicit requirements. Refactor across selected packages, then review every creation, edit and deletion before installing the accepted change.
Explore ConquistarFree beta for Windows 11 x64.
Bring Docker Desktop and your own Codex access.
Public packages first. Offline checks after.
conq prepare-dependencies typescript --source PROJECT --image IMAGE --output DEPS.jsonCommand shown with project and image placeholders.
A separate, networked preparation step fetches supported public locked dependencies. Candidate checks use the admitted image offline.
A first month on the network.
$20 / month
$40
included inference credit
≈ 4,000 QAIS at the $0.01 modeling rate — not a market price.
The working plate.
$50 / month
$125
included inference credit
≈ 12,500 QAIS at the $0.01 modeling rate — not a market price.
For heavy sessions.
$100 / month
$250
included inference credit
≈ 25,000 QAIS at the $0.01 modeling rate — not a market price.