Green Compute is Bittensor subnet 110 (SN110), a project in the Inference and Compute lane. Green-energy GPUs, scored every epoch
A Bittensor subnet that rents out GPUs for AI inference with one hard condition: miners have to prove their electricity is renewable, every epoch, or they do not get paid.
It is an educational deep dive, not a recommendation to stake. For live price, pool depth, and performance, always open the live subnet page.
What is Green Compute (SN110)?
Green Compute is a decentralized GPU compute network running as subnet 110 on Bittensor. According to its website, it rents out GPU capacity and serves AI inference, and miners only qualify if they can prove the power running their hardware comes from renewable sources like solar, hydro, wind, geothermal, or biogas.
In plain English: It is like renting cloud GPUs the way you would from RunPod or Lambda, except every machine on the network has to run on verified clean power.
Centralized reference points: AWS, CoreWeave, Lambda, RunPod
How miners and validators work
Like every Bittensor subnet, SN110 is a two-sided market: miners produce work, validators score it, and emissions follow those weights under the network’s dTAO economics.
- Miners run GPUs (the site names RTX 4090 and 5090 cards), serve real inference requests, and have to prove their power is clean through carbon registries and hardware-location attestation that is re-checked each epoch. They earn TAO plus rental fees, paid per epoch, on a "no proof, no payout" basis.
- Validators score the miners each epoch. The site states that validators score miner work but does not publicly spell out exactly what they verify.
Mechanism and design
Per the project's website, miners register GPU hardware (two SKUs, RTX 4090 and 5090) to subnet 110 and serve inference behind an OpenAI-compatible API. Before they can earn, they have to verify their energy source through carbon registries and hardware-location proofs, and that attestation is re-run each epoch via remote attestation. Validators score miners on their work, and payouts in TAO plus rental fees are settled each epoch, with the site framing it as "no proof, no payout." The website advertises indicative pricing from $0.40 per GPU-hour for a 4090 and $0.70 for a 5090.
A few honest caveats. The implementation is not independently verifiable from here: there is no public code repository registered to the slot on-chain, the site lists no named team, and the product, pricing, and performance claims all come from the project's own marketing site. The validator side of the mechanism, specifically what validators check beyond "scoring," is not publicly documented. Treat the product details above as the team's description of what they are building, not as independently confirmed fact.
On-chain, the slot is over a year old and active, with an emission share near 2.6% and pool depth of roughly 5,558 TAO backing the alpha token. The strong recent price move sits against thin public evidence of usage, so the open question is whether real inference demand shows up on the network.
Why the project exists
- The problem it solves: AI inference is expensive to run and increasingly energy-hungry, and data-center power draw is a growing concern. Green Compute's pitch is to tie payouts to electricity that is provably renewable rather than treating "green" as a marketing badge.
- The opportunity: It gives renewable-energy operators a way to monetize spare GPU capacity, and gives buyers an OpenAI-compatible API served on green hardware (both per the project's website).
- The Bittensor advantage: The incentive layer pays miners in TAO for serving inference, and the green-energy proof is enforced as a condition of payment each epoch rather than self-reported once at signup.
- Traction signals: Be honest here. The subnet carries an on-chain emission share around 2.6%, and its alpha price has roughly doubled over the past 30 days. But the project is early by its own roadmap, which dates a testnet to February 2026 and mainnet on subnet 110 to April 2026. No public code repository is registered on-chain, and there is no independent social or usage signal about the subnet yet.
What stakers should check before buying alpha
Subnet storytelling is cheap. On-chain and product checks are not. Use the same framework as our subnet metrics guide:
| Check | Why |
|---|---|
| Real product / docs / repo | Separates shipping teams from narrative-only slots |
| Pool depth (TAO) | Your exit door: thin pools amplify slippage |
| Net flows (7d / 30d) | Under flow-based emissions, capital votes with its feet |
| Emission share | Whether the market is funding this subnet right now |
| Ownership concentration | Large holders can move a small alpha market |
| Immunity / deregistration | Post-immunity weak slots can lose the race (see immunity guide) |
Pull the live numbers on Green Compute · SN110 rather than trusting any static screenshot in a blog post.
Risk factors
- Execution: Green Compute is early by its own published roadmap (testnet February 2026, mainnet on subnet 110 April 2026). Building a green-verified GPU network, the attestation pipeline, and real inference demand at once is unproven.
- Single-source documentation: Product, pricing, and performance claims rest on the project's own website. No public code repository is registered on-chain and no team is named, so the implementation cannot be independently verified here.
- Traction unproven: There is no independent social or usage signal about the subnet yet, and on-chain telemetry currently shows no active miners. Whether demand materializes is the central unknown.
- Liquidity: Pool depth around 5,558 TAO is modest, so larger stakes will face meaningful slippage on entry and exit.
- Competition: Multiple Bittensor subnets target compute and inference, and centralized clouds compete on price and scale. The green-energy requirement is the differentiator that has to pay off.
How this fits a TAO portfolio
Many holders keep a root vs subnet split: root for calmer TAO exposure, subnets like Green Compute for thematic upside (and higher variance). If you use TaoScale, you can size SN110 from your wallet without handing over custody. See non-custodial staking.
- Live market: SN110 on TaoScale
- Browse and stake: /invest
- More subnet explainers: /blog/subnets
FAQ
What does Green Compute do on Bittensor?
It is like renting cloud GPUs the way you would from RunPod or Lambda, except every machine on the network has to run on verified clean power.
What is the SN110 alpha token?
When you stake TAO into subnet 110, you receive that subnet’s alpha. Its price versus TAO moves with the subnet’s pool (see what is dTAO).
How do I research Green Compute before staking?
Read the mechanism above, open the official repo/docs when available, then verify pool depth, flows, and activity on SN110. Cross-check with the metrics checklist.
Is this financial advice?
No. TaoScale publishes educational explainers and live tools. Decisions to stake are yours alone.
Not financial advice. Educational overview of Green Compute (SN110) compiled from public project materials and ecosystem documentation. Mechanisms and market structure change, so verify live data and official sources before acting.
