Developer transparency

FinanceGPT ML Service

Public architecture, capabilities and verified runtime status for FinanceGPT quantitative machine learning.

API documentationOpenAPIStatus JSONTechnical report
Runtime status
Verification required
No verified runtime manifest has been published yet.
Numerical runtime
Verification pending
Numerical imports must pass before ML is considered ready.
Model runtime
Verification pending
Execution hardware is verified separately.
Inference
Pending pass
Operational status requires deterministic inference verification.
Verification boundary: FinanceGPT does not claim this ML runtime is operational until numerical imports, service startup and model inference smoke tests have all passed. The most recent internal installation evidence identified a NumPy CPU-instruction compatibility issue that must be remediated on the deployment host.

Architecture

FinanceGPT API / Quant workflows ↓ Governed ML gateway ↓ Model registry + version policy ↓ FastAPI ML service ├─ statistical / anomaly models ├─ generative financial models ├─ time-series models └─ evaluation & lineage ↓ Evidence + model/version metadata

Readiness gates

NumPy numerical importPending
PyTorch importPending
Service startup / healthPending
Model inference smoke testPending
Execution deviceUNKNOWN

Capabilities

ML complements deterministic FinanceGPT quantitative models; availability depends on the verified runtime.
Anomaly detectionGenerative scenariosVAE / conditional VAETime-series generationModel evaluationVersioned inferenceEvidence lineage

Public disclosure policy

Software capability and deployment readiness are reported separately. GPU runtime packages do not prove that physical GPU acceleration is present, and no operational claim is made from package installation alone.

Internal host names, filesystem paths, credentials, package-install commands and private infrastructure details are intentionally excluded from this public report.