Models with the largest context windows, ranked by tokens of context — for long documents, RAG and whole-codebase workloads.
15 models ranked
Rankings use live pricing from TestKey plus HotON.ai's composite efficiency score. Pricing is real and updated daily; the efficiency score is a modeled composite of real inputs — see our methodology. This list reflects 15 tracked models for this use case.
Pricing is real (via TestKey, daily). Performance metrics are modeled estimates — not financial advice.