Hosted inference APIs (15)
Hosted inference APIs: where a model can be rented by the token rather than run on owned hardware. Price fields carry one representative model named in price_basis — the catalog is per-model, so read the basis before quoting a number.
Novita AI serverless endpoints
other$0.135/$0.4 per Mtok45.0 tok/sTTFT 1770.0 msservices/svc-novita
contestedmedium · 0.78
SambaNova Cloud inference
sambanova$0.6/$1.2 per Mtok286.3 tok/sTTFT 2060.0 msservices/svc-sambanova-cloud
contestedmedium · 0.82
Amazon Bedrock managed inference
aws-bedrockservices/svc-aws-bedrock
verifiedmedium · 0.70
Azure OpenAI Service / Azure AI Foundry inference
azure$2.0/$8.0 per Mtok132.2 tok/sTTFT 2280.0 msservices/svc-azure-ai-foundry
verifiedmedium · 0.80
Cerebras Cloud inference (shared)
cerebras$0.35/$0.75 per Mtok3000.0 tok/sservices/svc-cerebras-cloud
verifiedhigh · 0.88
DeepInfra serverless inference
deepinfra$0.1/$0.32 per Mtok14.9 tok/sTTFT 2140.0 msservices/svc-deepinfra
verifiedhigh · 0.90
Fireworks AI serverless inference
fireworks$0.15/$0.6 per Mtokservices/svc-fireworks-serverless
verifiedmedium · 0.80
Google Cloud Vertex AI generative inference
gcp-vertex$0.72/$0.72 per Mtok153.6 tok/sTTFT 680.0 msservices/svc-gcp-vertex
verifiedhigh · 0.88
GroqCloud serverless inference
groq$0.15/$0.6 per Mtok305.7 tok/sTTFT 1000.0 msservices/svc-groqcloud
verifiedmedium · 0.80
Hyperbolic AI token inference API
other$0.12/$0.3 per Mtokservices/svc-hyperbolic
verifiedmedium · 0.72
Nebius Token Factory managed inference
other$0.15/$0.6 per Mtokservices/svc-nebius-token-factory
verifiedhigh · 0.85
OpenRouter hosted API router
openrouter$0.59/$0.79 per Mtok191.0 tok/sTTFT 520.0 msservices/svc-openrouter
verifiedhigh · 0.85
Scaleway Generative APIs (serverless, Paris)
other$1.0128/$1.0128 per Mtok83.0 tok/sTTFT 1330.0 msservices/svc-scaleway-generative-apis
verifiedhigh · 0.85
Together AI serverless inference
together$1.04/$1.04 per Mtok79.0 tok/sTTFT 1500.0 msservices/svc-together-serverless
verifiedhigh · 0.85
Lambda Inference API (hosted chat completions)
otherservices/svc-lambda-chat
deprecatedlow · 0.60