Why is the Claude Sonnet model more credits than the GPT-4o model
Why is the Claude Sonnet model more credits than the GPT-4o model
At a glance
The community member is curious why the Claude Sonnet model in LlmaParse costs more credits than the GPT-4o model. The comments suggest that this is because Anthropic has more expensive output token costs and input image costs, and that inputting images in the Sonnet pipeline consumes more tokens than in the GPT-4o model. However, there is no explicitly marked answer in the provided information.
The Sonnet pipeline is currently costing us more than the GPT-4o one for different reasons, one being inputing images in Sonnet consumes way more tokens than GPT-4o.