# How to Earn with io.net: Worker Setup and Real Payouts > How to supply your GPU to io.net as a Worker, what the hardware bar is, and why most worker income is block-reward emissions rather than client fees. The honest payout picture, plus the withdrawal caveats io.net's own docs flag. - Published: 2026-07-30 ยท Updated: 2026-09-24 - Canonical: https://ownyourmind.ai/use/how-to-earn-with-io-net/ - Cite as: Own Your Mind (ownyourmind.ai), licence CC BY 4.0. Link the canonical URL. Charts and interactive widgets are omitted here; project data is at https://ownyourmind.ai/api/scores.json. --- ## The honest earnings picture io.net registered a huge GPU supply and comparatively little paid demand, and that imbalance decides what a worker earns. Independent forum reports from GPU owners describe workers sitting idle most of the time and picking up few paid jobs, earning mainly from block rewards rather than paid compute. Over the first fortnight of June 2026, io.net's explorer showed client IO purchases at about 2.6% of supplier payouts. Since then its cumulative counters have climbed to about 5.8M IO of purchases against 9.8M IO of epoch rewards (September 2026), so demand is no longer negligible on io.net's own figures. The purchases settle off-chain and arrive in lumps, so they remain io.net's number; the per-epoch burns they fund do reconcile on-chain. Either way, what lands in a worker's wallet is epoch-reward emissions in an inflationary token, not a wage paid directly by AI customers. There's a second catch the marketing skips: getting paid out. By io.net's own documentation, worker job earnings can't yet be withdrawn, USDC withdrawals run through a manual support ticket, and automated payouts aren't enabled. Treat io.net worker income as accrued IO you'll manage later, not cash landing in your account each week. ## What you need The io.net Worker client runs on Windows, macOS and Linux, and the hardware bar is lower than a full Akash provider, which is why it sits at intermediate rather than advanced. - **GPU:** an NVIDIA GeForce RTX 3000-series card or above on Windows/Linux, or Apple M1/M2/M3 on macOS. AMD GPUs and CPUs aren't supported. - **Software:** Docker and current NVIDIA drivers installed. The Worker binary itself is closed-source, so you're trusting a pre-compiled executable. - **Connectivity:** at least 100 Mbps download and 75 Mbps upload. Multi-GPU rigs need 8x PCIe lane connectivity per card to pass verification cleanly. - **Stake:** a minimum amount of IO staked as collateral against your device (io.net doesn't publish the exact figure). If you own a gaming PC with a recent NVIDIA card, you already clear the bar. The friction is software setup and the token stake, not exotic hardware. ## How you get paid A worker earns two ways: job payments when the network assigns your GPU to a customer cluster, and hourly block rewards for being online and verified. In practice the second dominates, because paid job flow is thin relative to registered supply. Payouts settle in IO on Solana, so the same currency point from any token network applies. Selling IO on receipt turns the emissions into something close to a cash wage; holding IO is a separate bet on a token that is net inflationary with unlocks running to 2028. A worker who holds should book it as a position, not as compute income. Verification is continuous. The Worker passes Proof-of-Work benchmarks and a time-lock check, and your staked IO is the collateral that keeps you honest. Fall offline or fail verification and you stop earning block rewards, which is why uptime and a stable connection matter as much as the card itself. ## Setting up a worker The exact steps track io.net's docs, which change with each client release, so use them as the source of truth. The shape is straightforward. 1. **Prepare the machine.** Install Docker and current NVIDIA drivers, and confirm your GPU and internet speed clear the minimums above. 2. **Create a worker.** In the io.net Worker area, create a device, choose the connectivity tier, and generate the install command for your operating system. 3. **Run the binary.** Paste the command to install and start the IO Worker. It registers your device, runs the Proof-of-Work benchmark, and joins the supply pool once verified. 4. **Stake and connect a wallet.** Connect a Solana wallet, stake the required IO against the device, and confirm the device shows as active in your dashboard. 5. **Monitor earnings.** Watch block rewards accrue and job assignments arrive. Remember the withdrawal caveats above when you plan to realise any of it. ## Is it worth it? For someone with an idle NVIDIA gaming rig, io.net is a low-bar way to earn IO at the margin, as long as you go in understanding that most of it is emissions and that cashing out has friction today. It suits tinkerers comfortable with Docker and managing a volatile token more than anyone after predictable income. For predictable, cash-denominated GPU income, a cash marketplace is the cleaner choice, and Akash pays providers from actual client fees rather than emissions (though its demand is thin too). For how io.net's verifiable revenue compares with Render, Akash and Aethir, see [RENDER vs AKT vs IO vs ATH](https://ownyourmind.ai/tokenomics/render-vs-akash-vs-ionet/). For the same decision across cash marketplaces and token networks, see [what compute actually pays](https://ownyourmind.ai/use/gpu-provider-income/) and the sibling guide on [running an Akash provider](https://ownyourmind.ai/use/how-to-run-an-akash-provider/).