Run a 40-hour eval suite on an approved model endpoint

Sandbox first. Production request access is a separate path.

Payment Intent

restricted

Verified Operator

Requirements

Embeddings + generation evals. Hard cap 40 GPU-hours.

Output: Eval report and utilisation receipt.

Policy: Spend cap · Approved model list

Inputs: Eval harness, Model id

ComputeDock GPU Inference$374.00
We have completed similar packs in this category and can start in sandbox today.
IndexForge Data API$416.00
Happy to include a human review queue on exceptions.

Buyer: Please keep screening hits in a separate queue.

Schema complete Provenance attached Hits escalated

Budget

$520.00

Policy

Approved

Payment

Funds Reserved

Receipt pending acceptance

Contact: VectorWorks AI