Inference Engines & InfraFreemium
TokenRoute
The core loop is: feed your real prompt contracts into benchmark packs, run them against candidate models (local Ollama, private endpoints…
VerifiedFreemium
View TokenRouteCategory
Model Hosting APIs
Hosted inference APIs (Replicate, Together, Fireworks).
Inference Engines & InfraPaid
ParseRail
ParseRail exposes 39 single-call endpoints covering document parsing, PII redaction, commerce ops like chargeback analysis and PO matching…
Verified
View ParseRail
Inference Engines & InfraFreemium
Milliseconds.ai
The API exposes six endpoints — yes/no answers, classification, rating, open-answer extraction, entity identification, and field extraction…
VerifiedFreemium
View Milliseconds.ai
Inference Engines & InfraPaid
TrustedRouter
The tool is a paid API gateway that sits between your existing OpenAI SDK calls and 600+ models across 90+ providers, routing traffic…
Verified
View TrustedRouter
Inference Engines & InfraPaid
1endpoint
The gateway accepts OpenAI and Anthropic request shapes natively, so existing client code points at one base URL and keeps working. Prompt…
Verified
View 1endpoint
Inference Engines & InfraPaid
ApiModels
ApiModels routes calls to text, image, video, and audio models from OpenAI, Anthropic, Google, ByteDance, ElevenLabs, and others through…
Verified
View ApiModels
Inference Engines & InfraPaid
Prizm
Prizm routes requests to Veo 3.1, Kling, Seedance, Nano Banana, FLUX, and ElevenLabs through a single API key, drawing costs from a shared…
Verified
View Prizm
Inference Engines & Infra
WitFlow
WitFlow centralizes model routing, so requests go through one control plane instead of scattered direct integrations — which means cost…
Verified
View WitFlow
Inference Engines & InfraFree
Agent Router
Agent Router is a gateway that sits in front of multiple LLM providers and exposes a single OpenAI-compatible endpoint, so any framework…
Verified
View Agent Router
Inference Engines & InfraFreemium
OmniRoute
The vendor describes OmniRoute as a self-hosted gateway that exposes a single OpenAI-compatible endpoint at localhost:20128/v1 and routes…
VerifiedOSSFreemium
View OmniRoute