daitchain

Run a compute host

A compute host runs the dait-host daemon inside a confidential VM, serves AI inference jobs from an attested TEE, signs receipts with a TEE-bound key, and posts those receipts back to the chain for PoUW credit. Hosts earn the 60 percent share of block rewards plus tenant lease payments.

Hardware floor is real. Confidential Computing for AI requires Hopper-or-newer NVIDIA cards with CC mode enabled, plus a CPU with TDX or SEV-SNP. Pre-Hopper cards (A100, L40S) cannot produce a NRAS quote and are not supported. Honest entry ticket: ~$22k-$44k for a single-GPU node.

Hardware

ComponentMinimumNotes
CPUIntel Xeon Sapphire Rapids+ with TDX, or AMD EPYC Genoa+ with SEV-SNPOlder CPUs cannot attest. TDX requires 4th-gen Xeon or newer; SEV-SNP requires EPYC 7003 or newer.
RAM256 GB ECCThe CC VM eats one full memory channel. 128 GB is too tight for 70B-class models.
GPU1x H100 80 GB SXM, or H200, or B200, or GB200Hopper or newer. CC mode must be enabled in firmware.
NIC25 GbE minimum, 100 GbE recommendedState channel servers see bursty token streams.
Storage2 TB NVMe Gen4Model weights, receipt log, attestation cache.
BMCOut-of-band on a separate VLANRequired for evidence-of-physical-control audits.

Stake

3,690 DAIT per registered GPU at the Node tier. 36,900 DAIT per multi-GPU rack at the Supernode tier (1.25x reward multiplier, priority routing). See x/staking_tiers for the full tier table and slashing rules.

Software stack

+----------------------------- physical host -----------------------------+
|  host OS (Ubuntu 24.04 or RHEL 9)                                       |
|    +-- KVM / QEMU                                                       |
|         +-- confidential VM (TDX or SEV-SNP guest)                      |
|              +-- dait-host daemon                                       |
|              +-- vLLM worker                                            |
|              +-- NVIDIA driver in CC mode                               |
+-------------------------------------------------------------------------+

The daemon expects to run inside the confidential VM, not on the host OS. /dev/tdx-guest or /dev/sev-guest must be exposed inside the VM, and the NVIDIA Local Verifier SDK must be installed in the same image.

Setup

# 1. Provision the confidential VM image (per-vendor; see vendor docs)

# 2. Inside the VM, install dait-host
pip install dait-host

# 3. Configure
export DAIT_CHAIN_RPC=https://rpc.daitchain.io:443
export DAIT_HOST_KEY_PATH=/var/lib/dait-host/operator.key
export DAIT_TEE_KIND=nvidia-nras   # or intel-tdx, amd-sev-snp
dait-host init

# 4. Register on-chain
dait-host register --gpu-uuid GPU-XXXXXX-XXXX-XXXX

# 5. Submit attestation (rotates every Params.EpochNonceRotationBlocks)
dait-host attest

# 6. Start serving
dait-host serve --models qwen2.5-32b-instruct,llama3.1-8b-instruct

The receipt loop

  1. Tenant opens a state channel with the host (MsgOpenChannel).
  2. Tenant sends inference requests over HTTPS to the host's channel server.
  3. Host runs the request inside the TEE, produces a signed SPUR-IC receipt, returns the response with X-DAIT-Receipt and X-DAIT-State headers.
  4. Tenant verifies the receipt, counter-signs the new ChannelState, ships it back on the next request.
  5. Periodically the host calls MsgCommitReceipt, then MsgRevealReceipt after commit_reveal_window blocks. Reveal mints PoUW credits.
  6. On channel close (cooperative or challenge), the channel's spent_amount moves from tenant escrow to host balance.

Determinism rules

Receipts must be reproducible by an auditor. The reference daemon serves replayable: true jobs with --enforce-eager --seed=N mode in vLLM and writes response_hash over the canonicalized JSON output. Audits replay the receipt and reject any that do not produce a bit-exact match.

Slashing

See also