NewsOfficialXSep 30, 2026, 16:36 UTC

Subnet 47 is building verifiable GPU training on Bittensor.

Subnet 47 is building verifiable GPU training on Bittensor. H100 attestation. Sandboxed workloads. Hidden challenges. Verified checkpoints and throughput. 381 tests passed. The thesis: reward proven AI compute, not self-reported hardware. Still pre-alpha. github.com/forgenet47/gpuforge #GPUForge #Bittensor #AI #DePIN

Linkgithub.comgithub.com/forgenet47/gpuforge

Open on X ↗

SN47 GPUForge at a glance

SN47, GPUForge (α token צ), is one of the independent markets on the Bittensor network, each with its own α token and its own price. Its α currently trades at 0.002029 τ (≈ $0.61), against 1,074 τ of pool liquidity. On-chain, the best-delegated validator on it yields about 90.23% APR (simple) and 146.51% APY (compounded), emissions add 0.000026 τ to the pool each block, epochs run every 360 blocks. It was registered at block #7,340,355.

Chart, holders, conviction and every figure →

More from SN47 GPUForge

OfficialX · @GPUForce · 4d ago

🔥 GPUForge: big code update

OfficialX · @GPUForce · 7d ago

🔥 GPUForge code update 📦 Artifact integrity checks + resumable transfers 🖥️ Signed H100 capability reports 🔐 Scoped access hooks 🧪 New failure-case tests Pre-production. GPU claims remain self-reported. https://t.co/x9Le7SYeQB $TAO #Bittensor #DeAI https://t.co/Ridxv5jem1

OfficialX · @GPUForce · 8d ago

🖥️ Your GPU says the training job is done. What makes that claim believable? Imagine sending a training job to a remote machine. It reports completion and impressive throughput. Did it use the correct data? Run the specified training code? Does the result belong to this

OfficialX · @GPUForce · 9d ago

⚡ Introducing GPUForge- Subnet 47 A pre-alpha Bittensor subnet for verifiable GPU training. The idea is simple: Turn GPU capacity into AI training work that can be verified. 🏭 Publisher — creates immutable workloads & signs manifests ⚡ Miners — execute workloads on GPUs 🛡️ https://t.co/aHvEdpD4kF

Taosis indexes announcements, articles, posts and releases about Bittensor subnets from their own channels and keeps a record of each. The text above is the source’s; the figures are read from the chain by Taosis.