Compare
Pick any two models. See the difference.
Search the 1537-model catalog across 88 providers, then open a side-by-side breakdown — full provider matrix, specs, and a calculator that runs the same workload on each side.
Model A
Model B
Pick a model on both sides.
Or jump to a popular pair
Popular comparisons
Compare current models in the same capability tier
Start with current cross-provider pairs selected from the live catalog. These links match broadly comparable models instead of putting a frontier model against a budget endpoint simply because both are new.
Comparison guide
How to compare LLM API pricing
Separate input and output
Prompt-heavy retrieval workloads care most about input rates. Agents, code generation, and long-form responses can spend more on output tokens. Compare both before choosing.
Use the same workload
A lower per-token price does not always mean a lower bill. Enter the same request volume, prompt length, and response length for both models to estimate the real gap.
Check provider differences
The same model can have different prices across native APIs and inference providers. Review context limits, caching, batch discounts, and availability with the price.