SENDFU-operated inference with clear model provenance
AI inference, operated by SENDFU.
SENDFU operates the complete inference service through a unified OpenAI-compatible API and Singapore-based infrastructure; base-model ownership remains with each model's developer.
Built for dependable model access. A unified API, explicit model allowlist, authenticated access, streaming support, and clear service documentation.
Model servicesAvailable models are published through the SENDFU catalog, with separate documentation for SENDFU-developed model variants and controlled-access services.
Data principlesData handling terms and retention commitments must be evaluated against the applicable SENDFU service terms before use.
Data handling boundaries
Review Trust Center →Review the applicable SENDFU service terms before sending production data or relying on a retention commitment.
Explore SENDFU.
Product, API, access, trust, and company details live in dedicated pages.
Model routes for platform and enterprise applications.
SENDFU publishes authenticated, OpenAI-compatible chat-completions services for selected models, alongside separately documented SENDFU-developed model information. The public catalog is the current source for available model IDs.