InferenceList

Pricing explorer

Token rows use input and output per 1 million tokens. Media rows use a flat USD amount per unit.

Provider all AnthropicAssemblyAICartesiaCloudflare Workers AIDeepgramDeepSeekElevenLabsfalFireworks AIGoogleGroqIdeogramLuma AIMistral AIOpenAIReplicateRunwayStability AISunoTogether AIxAI Unit all character_audioimageminute_audiorequestsecond_videotoken Apply

Model	Provider	Unit	Input / 1M	Output / 1M	Flat	Confidence
luma/ray-2	luma	second_video	—	—	$0.1	seed