Skip to content
Small team, full backlog, zero orders dropped. Support replies are slower than we’d like. Read our status update → Zero orders dropped. Status → 📬 Check your spam folder — most of our replies land there. We do answer. Status update → 📬 Check your spam folder. Status →
1325 Rue Bergar, Laval, QC support@d-central.tech 1-855-753-9997 Mon-Fri: 8:30 - 18:00
Sovereign AI Workstation 24

Sovereign AI Workstation 24

Priced by quote

A 24 GB-class workstation that runs 30–32B models fast — a genuinely capable local coding assistant and document-RAG box. Ships with your chosen open model installed and running. The developer’s sovereign daily driver.

  • Condition: Refurbished
  • Availability: Built to order — priced by quote
  • Lead time: Hand-built to order in Montreal, Quebec. We keep inventory lean and build to your spec; your lead time is confirmed with the quote.
  • Returns: Made-to-order work is non-refundable once sourcing or assembly starts, except dead-on-arrival or defective units.
  • Warranty: Bench-tested and backed by D-Central's own warranty, as stated on this listing at purchase.
  • Support: Ships from Canada with D-Central mining hardware and repair support

Shipping, returns, and warranty details

Request a quote

Tell us your use case — we reply with a spec and a firm quote. Or call 1-855-753-9997.

Purchase details

  • Condition: refurbished

We accept Bitcoin and Lightning at checkout

We accept:
Shipping calculated at checkout
Ships from Canada
Bitcoin Accepted
Warranty terms shown before purchase
Ships from Canada
Secure Checkout
In-House Repair Experts

Save 3% when you pay in Bitcoin.

Shipping options and lead times are confirmed at checkout

Refurbished — bench-tested before it ships

Every refurbished unit passes the same bench it took to write our repair guides: visual inspection, full chip-count / pattern test, per-domain voltage check, thermal cycle, burn-in at rated spec, and proactive replacement of worn fans, thermal paste, and connectors. Backed by D-Central's own warranty.

See exactly what we test →

SKU: DC-AI-WS24 Category:

Questions about this product?

Talk to a mining expert → or call 1-855-753-9997

Description

For the developer or creator who wants a real local coding and reasoning box. Built around a 24 GB GPU — an RTX 5090, or a refurbished RTX 3090 (tagged as refurbished) for the best dollars-per-VRAM — it runs 30–32B open models comfortably and serves a private coding agent that never sends your code anywhere.

Arrives with the open stack installed and a model loaded for your workflow. Credit to the open-model and open-tooling communities; our value is the build, the sizing, and the power-and-heat work that keeps it fast and quiet.

Honest limits: a 24 GB box won’t run the largest models at full context; we’ll size to your real workload. Built to order, priced by quote. How a local coding agent compares.

You May Also Need

Related Products

Learn More About This Hardware

Bitcoin × AI

Pay-Per-Inference, No Account: The Sovereign Compute Loop

Sovereign compute means running AI inference on hardware you own and paying for it in Bitcoin over Lightning with no account. Here is the canonical definition of the sovereign compute loop — your hardware, your model, your money — and how L402 pay-per-inference closes it.

Read More on the Blog

Related products, repair, and setup paths

Last reviewed June 15, 2026.