InferIndex

InferIndex is an independent index of LLM API prices. Ask Claude which provider serves a model most cheaply, compare every provider's price for a model, look at how prices changed, or estimate what a workload will cost. - Cheapest offer for a model, with filters that matter in production: no training on your prompts, processing region, quantization, context length, tool calling, JSON mode and vision. - Every provider's current price for a model, direct providers and aggregators, with the conditions of each offer. - Price history, including the lab's official list prices back to 2023. - Cost estimates from your usage: prompt and output tokens, cached share and requests per day, including tiered and peak-hour pricing. Prices come from 70+ pricing sources and are re-checked every few hours. Stale offers, expired promotions and prices that aren't a real per-token rate are flagged and never presented as the cheapest.