The same model can be cheap for input-heavy RAG and expensive for output-heavy translation. Pick a workload shape to see the active catalog ranked by monthly cost.
Compare active models for a support chatbot handling 30M input and 6M output tokens per month, ranked by stored catalog cost.
Rank active models for a coding assistant with 20M input and 4M output tokens per month using stored catalog prices.
Compare active models for a retrieval-augmented generation pipeline at 50M input and 5M output tokens per month.
Rank active models for document summarization at 40M input and 8M output tokens per month.
Compare active models for agent workflows with retries and tool calls at 25M input and 10M output tokens per month.
Rank active models for translation workloads at 35M input and 35M output tokens per month.