Platform feature · Hardware-free pricing
Estimate: price hardware you do not have.
Score how your real workloads perform across unowned chip classes—with zero hardware access needed and nothing run.
How it works
Capture workload math
Inspect the operations, memory pressure, and communication your program performs.
Score chip profiles
Price candidate options across ORIQX chip-class profiles for CPUs, GPUs, TPUs, NPUs, and QPUs.
Surface four signals
Report predicted time, plus cost, energy, and accuracy wherever each class can be priced.
Support hard decisions
Compare unowned architectures objectively before reserving cloud capacity or buying hardware.
What you get
- Zero access to candidate hardware required and zero compute jobs executed.
- Objective price and runtime comparisons before procurement or cloud commitments.
- Covers CPUs, GPUs, TPUs, NPUs, and QPUs, with more classes added as needed.
When to use it & who it is for
Built for infrastructure architects, procurement leads, sovereignty planners, and security/threat evaluators.
What Estimate is used for
Buy or rent decisions
Decide which processor classes to buy or lease before committing capital.
Sovereignty design
Plan compute capacity you control around real workload math and budget constraints.
Evaluating threats
Judge what a capable adversary could compute across chip classes without owning their hardware.
Estimate without hardware access
Scores your program across CPU, GPU, TPU, NPU, and QPU chip classes without running jobs or accessing unowned hardware.
Estimate · Unowned hardware
ExamplePrice an AI training run on chips you do not own
Workload input
Input: model training & inference graph structure with memory footprint and attention stages.
ORIQX action
Estimate scores the workload across ORIQX chip-class profiles with zero hardware access.
More ORIQX platform features
Explore the rest of the ORIQX platform.
Workload Optimization
Takes your existing program unchanged, splits the work into compute blocks, and runs each block in parallel on the chip that suits it, fully managed or self-managed.
Explore OptimizePreflight
Lays out candidate ways to run a job across reachable providers and chips before launch—showing predicted time, cost, energy, and accuracy.
Explore PreflightAgent controls
Sits beneath existing agent frameworks so spawned retries, searches, and child simulations pass Preflight cost, speed, energy, and accuracy checks before nodes allocate.
See agentic compute