Arrakis vs Lambda
Lambda rents on-demand GPU virtual machines and clusters from its own data centers, billed by the minute on a weekly card invoice. Arrakis searches independent GPU markets for the lowest live price and rents on one prepaid balance, with spend caps and automatic shutdown for agents.
Side by side
As of 31 Oct 2026, 16:52 UTC, the lowest price for the NVIDIA B300 on Arrakis is $4.60 per GPU-hour, with 57 GPUs available on 8 machines.
| Arrakis | Lambda | |
|---|---|---|
| B300 per GPU-hour | $4.60/GPU-hr | Not listed: B300 only on contracts of 3 years or more |
| B200 per GPU-hour | $6.33/GPU-hr | $6.69 to $6.99 (B200 SXM6, 8 GPUs down to 1) |
| H100 per GPU-hour | $2.07/GPU-hr | $3.29 (H100 PCIe, one GPU); H100 SXM $3.99 to $4.29 |
| Billing | By the hour; a started hour counts in full | Per minute |
| How you pay | Prepaid balance: card, Link or Apple Pay | Weekly invoice to a credit card |
| Root SSH and any Docker image | Yes | Yes: a full VM, with sudo |
| Serverless endpoints, billed per request | No: each model gets a GPU of its own | No |
| Hosted models, billed per output | No: deploy any Hugging Face model yourself | No: its inference API is winding down |
| MCP server for AI agents | Yes, remote at /mcp | No official server found |
| Spend cap on each API key | Yes, on every token | No: API keys have full access |
| Automatic shutdown | Yes: every rental has a removal time | Not documented |
| Interruptible machines | Yes, where hosts offer them | No |
| Searches several GPU markets | Yes | No: its own cloud |
Facts about other providers come from their own pages, checked on October 31, 2026. Prices and features change: check them before you choose. Arrakis is not affiliated with them.
Which fits you
Lambda fits better
You want full virtual machines, multi-node GPU clusters, or a monthly or weekly invoice from one provider with its own data centers.
Arrakis fits better
You want prepaid spending you can't overrun, prices from independent markets, automatic shutdown on every machine, or an agent that rents GPUs on its own within a cap.
Moving to Arrakis
Your Docker image, SSH keys and scripts move as they are: Arrakis boots any public or private image with root SSH, and serves your own model with your own engine options. Start with one machine in the console, or point your agent at /mcp with a capped token.