Phase 1 — Apple Silicon · Mainnet Preview

The World's Idle Mac GPUs
Working For You

A decentralized marketplace for Apple M-series unified memory compute. Run Llama 3 70B on a Mac Mini M4 Pro for $0.07/hr. Earn HC credits while your Mac sleeps.

Mac Mini M4 Pro (48 GB) = two A100 80GB cards ($30,000+) for AI inference. No CUDA required. No cloud markup. Just Metal.

Available providers
Total GPU memory
Running jobs
Completed jobs

Apple Silicon — The Compute Advantage

Unified memory = GPU memory. No separate VRAM. A $1,400 Mac runs what needs $30,000 in NVIDIA cards.

Mac ModelUnified MemoryGPU CoresBest ForEst. Price
Mac Mini M416–24 GB10-coreSmall inference, simulation$0.03/hr
Mac Mini M4 ProPopular24–64 GB20-core30B LLM inference, SD$0.07/hr
Mac Studio M4 Max64–128 GB40-core70B LLM inference$0.12/hr
Mac Studio M4 Ultra128–192 GB80-coreFull 405B model shards$0.22/hr

Compared to AWS p4d.24xlarge (8× A100): $32.77/hr · No Apple Silicon support · 3× higher latency for LLM inference

Built for Real GPU Compute

Apple Silicon Advantage

Unified memory means a $1,400 Mac Mini M4 Pro runs Llama 3 70B — hardware that normally requires two $15,000 A100 cards.

Sandboxed Execution

Every job runs in a dedicated macOS sandbox-exec profile: restricted filesystem, localhost-only network, isolated OS user.

Zero-Copy GPU Access

MLX and PyTorch MPS use Metal directly. No CPU↔GPU copies, no driver headaches — full unified memory bandwidth from day one.

Decentralized Network

libp2p Kademlia DHT — no single point of failure. Providers connect peer-to-peer; the coordinator cluster is just a matchmaker.

Earn While You Sleep

Automatic idle detection via IOKit + CGSession. Your Mac starts accepting jobs only when screen is locked and GPU is idle.

HC Credit System

Off-chain HC credits for Phase 1. On-chain Arbitrum L2 token for trustless settlement in Phase 3 — no gas fees in the interim.

For Consumers

Run AI jobs in 3 steps

Submit Python scripts to the network. MLX, PyTorch MPS, ONNX CoreML — choose your runtime.

1
Install CLI
brew install hatch/tap/nm
2
Browse GPUs
nm gpu list --min-ram 48 --runtime mlx
3
Run your job
nm job submit --runtime mlx --ram 48 ./inference.py
Browse available GPUs
For Providers

Earn while your Mac sleeps

Install the agent on your Mac Mini or Mac Studio. Idle detection is automatic — jobs only run when your screen is locked.

1
Install agent
curl -fsSL https://raw.githubusercontent.com/wkang0223/neuralmesh/master/scripts/install-agent-macos.sh | bash
2
Set your price
nm provider config --idle-minutes 10 --floor-price 0.05
3
Start earning
nm provider start
Provider dashboard

Use from Python or CLI

One command to install — works with any Python project or Jupyter notebook.

Python SDKpip
pip install hatch

import hatch as nm

nm.configure(account_id="your-id")

job = nm.submit(
    script="./inference.py",
    runtime="mlx",
    ram_gb=48,
    hours=2,
)

for line in job.stream_logs():
    print(line, end="")
CLI (Homebrew)brew
# Install
brew install hatch/tap/nm

# Browse available Macs
nm gpu list --min-ram 48 --runtime mlx

# Submit a job
nm job submit \
  --runtime mlx \
  --ram 48 \
  --hours 2 \
  ./llama_inference.py

# Stream logs
nm job logs <job-id> --follow

How Credits & Payments Work

Transparent, real-time accounting. Every HC is tracked — no hidden fees, no surprises.

$1 = 1 HC
Deposit

Pay via Stripe or wire HC from Solana / Arbitrum. Credits appear instantly.

max budget held
Escrow

When a job is matched, your max budget is moved to escrow — safe, not spent.

price/hr × time
GPU Runs

The provider's GPU executes your job. Metered to the second.

auto on complete
Settlement

92% to provider · 8% platform fee · unused budget refunded to you.

anytime
Withdraw

Providers cash out to their Solana or Arbitrum wallet. No lock-up.

Example: 4-hour job at 0.10 HC/hr = 0.40 HC total
0.368 HCprovider (92%)
0.032 HCplatform (8%)
If the job finishes early (e.g. 2 hrs instead of 4), you only pay for 2 hrs. The remaining 0.20 HC is automatically refunded from escrow.
Phase 1 (now): Off-chain HC ledger (PostgreSQL)
Phase 2: On-chain SPL token (Solana) + ERC-20 (Arbitrum)
Phase 3: ZK compute proofs — trustless settlement

Roadmap

Phase 1Apple Silicon MVPIn Progress
Mac Mini M4 / Mac Studio providersMLX · PyTorch MPS · ONNX CoreMLOff-chain HC creditsmacOS sandbox isolation
Phase 2Hardening + Migration
DMTCP job checkpointing + migrationmacOS Virtualization.frameworkReputation systemNext.js dashboard v1
Phase 3On-chain Token
HC ERC-20 on Arbitrum L2Trustless escrow smart contractsZK Groth16 compute proofsProvider staking
Phase 4NVIDIA + AMD + Intel Arc
Linux agent (nvml · rocm_smi · Level Zero)Docker + NVIDIA Container ToolkitFirecracker MicroVMsApache TVM cross-platform compiler