Rent a whole GPU by the second.
Stop paying for the other 59 minutes.
GPU pods in Saudi Arabia: root, SSH and Jupyter in about a minute, billed per second in Riyal — and the unused time is refunded the instant you stop. When you don't need the whole card, the same mesh answers as an OpenAI-compatible API.
fig. 01 — whole-GPU pod · dedicated siliconA whole GPU. Root, Jupyter, SSH. In about a minute.
Not a slice, not a queue — the entire card is yours with a pinned driver. Launch from the console or one API call, extend without a restart, and stop whenever you want: the unused seconds come straight back to your wallet.
- Root + Jupyter over TLS + SSH · persistent /workspace volumes
- Prepaid per GPU-second in SAR · prorated refund on early stop
- Idempotent money routes · machine-readable 402 for agents
$ curl -X POST https://api.dcp.sa/api/pods \
-H "Authorization: Bearer $DCP_KEY" \
-d '{"gpu_type":"H100","duration_minutes":60}'
201 { "access_url": "https://…/jupyter", "ssh_command": "ssh …" }
fig. 02 — token streams · api.dcp.sa/v1Point your SDK at the Kingdom.
One line changes: base_url. Chat, embeddings and rerank served from KSA-resident GPUs, with streaming, function calling and JSON mode. An Arabic-first, open-source model lineup — frontier models stay off unless you opt in.
from openai import OpenAI
client = OpenAI(
base_url="https://api.dcp.sa/v1", # ← the only change
api_key=os.environ["DCP_API_KEY"],
)- Per-1M-token rates by class: 5 → 400 halala · shown before you commit
- GET /v1/models carries a live available flag — no stale catalog
Everything above runs inside Saudi Arabia.
Data, models, storage and the control plane stay in the Kingdom, under Saudi law — not as a policy promise, but because the hardware is here. Cross-border frontier models exist only behind an explicit per-tenant switch.
One curl. Whole GPU.
curl -X POST https://api.dcp.sa/api/pods \
-H "Authorization: Bearer $DCP_KEY" \
-d '{"gpu_type":"H100","duration_minutes":60}'Root, Jupyter and SSH come back in the response. Extend or stop it the same way.
Change one line. Keep your code.
client = OpenAI( base_url="https://api.dcp.sa/v1", api_key=DCP_KEY)
Chat, embeddings and rerank from in-Kingdom GPUs — streaming, function calling, JSON mode.
No terminal? The console.
Pick a GPU, click launch, and Jupyter opens in a tab. Wallet, invoices and usage live in the same place. The playground answers before you ever create a key.
Agents are first-class.
npx -y github:dhnpmp-tech/dcp-mcp
An agent self-registers, gets a real key plus a 20 SAR trial, and rents GPUs — zero humans involved.
DCP-Agent. Live for SMB.
The Arabic AI agent for Saudi small & mid-size businesses — already in production at agents.dcp.sa. A free personal version for every Saudi is coming.
An agent can rent a GPU by itself.
Self-register in one call, get a real key plus a 20 SAR trial credit, then rent pods and run inference through the official MCP server — no email click, no human.
npx -y github:dhnpmp-tech/dcp-mcp
fig. 03 — a provider node · somewhere in the KingdomYour idle GPU. Paid in Riyal.
One 4 MB app, packaged natively for every desk: a signed .exe for Windows, a universal .dmg for macOS, .AppImage and .deb for Linux. Run it once — it detects your card, installs the inference engine, downloads a model, verifies real throughput, joins the self-hosted WireGuard mesh, and keeps itself up to date. No port forwarding, no DevOps.
- You keep 75% of every Riyal your rig earns · paid out in SAR
- Native installers + signed auto-updates · one codebase, three platforms
- Paid only while your machine is verified and actually serving
How do I rent an H100 (or other GPU) on demand on DCP?
Sign up for a DCP renter account at dcp.sa, fund your wallet in Saudi Riyal, then launch a pod from the console or via the API — POST https://api.dcp.sa/api/pods with a Bearer renter key. You get a whole NVIDIA GPU (H200, H100, A100, L40S, RTX 5090 or RTX 4090) with root, Jupyter over TLS and SSH in about a minute. Billing is prepaid per GPU-second in SAR, with a prorated refund when you stop early.
Is DCP an OpenAI-compatible inference API?
Yes. DCP exposes an OpenAI-compatible API at https://api.dcp.sa/v1 (POST /v1/chat/completions, GET /v1/models). Point any OpenAI SDK at it by setting base_url to https://api.dcp.sa/v1 and using your DCP renter key as the Bearer token — no code rewrite needed. Inference is billed per token in Saudi Riyal.
Can an AI agent rent a GPU on DCP via MCP?
Yes. DCP ships an official Model Context Protocol (MCP) server. An MCP-capable agent (such as Claude) can list models, run inference, list available GPU types, create and extend GPU pods, rent storage volumes, and check wallet balance through tool calls. See dcp.sa/docs for the MCP setup and tool reference.
What is sovereign / in-Kingdom AI compute in Saudi Arabia?
Sovereign AI compute means your data, the models, the storage, and the control plane all stay inside Saudi Arabia, under Saudi law. DCP runs on Saudi-owned hardware in the Kingdom with full PDPL data-residency compliance, so prompts and answers never leave the country unless a tenant explicitly opts in to cross-border frontier models.
How much does it cost to rent a GPU on DCP?
GPU rental is billed prepaid per GPU-second in Saudi Riyal, cost-plus from the live market. On-demand types and indicative hourly rates: NVIDIA RTX 4090 from about 3.62 SAR/hr, RTX 5090 from 5.2 SAR/hr, L40S from 5.2 SAR/hr, A100 (80 GB) from 7.3 SAR/hr, H100 (80 GB) from 17.27 SAR/hr, and H200 (141 GB) from 23.05 SAR/hr. The native in-Kingdom RTX 3090 is 2.5 SAR/hr. New renter accounts start with 100 SAR of credit and no card is required to begin.
Where does my data live when I use DCP?
Inside Saudi Arabia. Inference, GPU pods, agents, and persistent storage volumes all run on in-Kingdom, Saudi-owned hardware under PDPL data-residency rules. Cross-border frontier models are available only by explicit per-tenant opt-in.
Sovereign Arabic AI. Run it.
Eight minutes from this page to a ready renter workspace. First inference is enabled by the catalog only when a verified serving model is online. No procurement. No data-egress conversation. No flat GPU rental.