VoltageGPU evidence bundle: single-GPU RTX PRO 6000 Blackwell Server Edition, Intel TDX guest
Captured 2026-09-17 00:00 UTC from inside a VoltageGPU Confidential VM (SKU rtx6000b-small).

What this shows
  Both proofs, generated by the tenant inside the VM, on a challenge the tenant chose:
  1. An Intel TDX quote (v4, TEE 0x81, 4 940 bytes) whose report_data is the SHA-512 of
     the manifest, verified offline up to the pinned Intel SGX Root CA, with Intel PCS
     collateral (TCB info, QE identity, CRLs) embedded. Platform TCB UpToDate.
  2. NVIDIA GPU attestation tokens from NRAS whose eat_nonce is the SHA-256 of the same
     manifest: measurements ok, secure boot on, debug off, report signature and chain
     validated. GPU-0 GB20X, driver 595.71.05, vbios 98.02.9E.00.01.

Environment read inside the guest (conf-compute.txt, gpu.txt, kernel.txt)
  systemd-detect-virt: kvm, /dev/tdx_guest present, kernel 6.8.0-110-generic
  nvidia-smi conf-compute -q: CC State ON, Multi-GPU Mode None, CPU CC INTEL TDX,
  GPU CC Capable, CC GPUs Ready State Ready
  Protected memory 99 463 296 KiB, unprotected 0 KiB

Why it matters
  Until this capture our product page said "Intel TDX quote only, GPU attestation not
  verified on this SKU" and our older notes said the RTX 6000B read CC OFF. That was
  stale. This is the third single-GPU SKU with both proofs, after H200 (2026-09-04) and
  H100 (2026-09-16), and the cheapest of the three.

Files
  manifest.json      the workload manifest with the random challenge (voltage-verify manifest --challenge auto)
  bundle.json        the evidence bundle (voltage-verify attest --mode single-gpu, run as root in the VM)
  verify.txt         full output of voltage-verify verify bundle.json, RESULT: VERIFIED
  conf-compute.txt   nvidia-smi conf-compute -q and -gm
  gpu.txt            nvidia-smi name, driver, vbios, memory
  kernel.txt         uname -r, /dev/tdx_guest, systemd-detect-virt
  SHA256SUMS         checksums computed inside the VM before download

Reproduce (as it actually runs on the VM image, which ships without pip)
  curl -sS https://bootstrap.pypa.io/get-pip.py | python3 - --user --break-system-packages
  python3 -m pip install --user --break-system-packages "voltage-verify[attest]"
  export PATH=$HOME/.local/bin:$PATH
  voltage-verify manifest --challenge auto -o manifest.json
  sudo env PATH=$PATH PYTHONPATH=$(python3 -c 'import site;print(site.getusersitepackages())') \
    voltage-verify attest --manifest manifest.json --mode single-gpu -o bundle.json
  voltage-verify verify bundle.json --challenge <the challenge printed by the manifest step>
  Then copy bundle.json off the VM and run the same verify on your own machine
  (pip install voltage-verify, https://github.com/Jabsama/voltage-verify, MIT), with
  --offline if you want no network at all. The NRAS tokens are signed by NVIDIA and the
  TDX quote by Intel's chain; a wrong --challenge makes verify answer NOT VERIFIED.
  A second bundle, produced the same day through the public API alone (deploy by API
  key, SSH, attest, stop), is in ../rtx6000b-api-2026-09-17/.

Limits, stated plainly
  This is one GPU. On 8-GPU nodes NVIDIA runs Protected PCIe mode and the NVSwitch fabric
  is not attestable from inside a TDX guest (see ../nvswitch-2026-09-16/). NVIDIA's CLI
  verifier fails from inside the guest because OCSP is unreachable; the Python SDK in
  remote mode works because those checks run on NVIDIA's side.

Checksums (sha256, computed in the VM)
  d793286612494ec22dca3f0d98607efc1b04ec86103d7c57a181f2889171ef41  bundle.json
  e6e145dcaae1dca0e50787aef77e08ff1887d9a0cac041adc9110f826f5a28dc  manifest.json
  300fdc87eb1dcdb5b46e5b18443339732a5bc3d26baa6aba88af3343652ec0bb  verify.txt
  a29635e51e19c2391b0b55f1b8bbd9713df9354f1075a7547d85f2f4dcf0543f  conf-compute.txt
  bf8c6451fe67ebdf7431c470b5ecc2f6367f7eb9e03e3251bfed59f33aa4bf5e  gpu.txt
  55b60a64b1b30af39dff7731e9f3be63595e13a293791d17f5514d4392c356d1  kernel.txt