Best Models by Use Case

Data-driven rankings of large language models and decision models for specific needs — from the cheapest APIs to the best calibrated decision models. Each list is built from live pricing, specs, and benchmark data, updated regularly.

LLMs

Large language models ranked by price, context window, open weights and benchmark scores.

Cheapest LLMs by Price per Token

The most affordable large language models ranked by output price per million tokens. Compare the lowest-cost LLM APIs by input and output cost.

LLMs with the Longest Context Window

Large language models ranked by context window size. Find the LLMs that can process the most tokens — long documents, large codebases, and extended conversations.

Best Open-Weight LLMs

Open-weight large language models you can self-host. Ranked open LLMs by context window and price, with openly available model weights.

Best LLMs for Coding

Large language models ranked by coding benchmark performance. Find the best LLMs for code generation, debugging, and software engineering tasks.

Cheapest Frontier LLMs

The most affordable frontier (state-of-the-art) large language models, ranked by output price per million tokens. Get top-tier capability at the lowest cost.

Decision models

Decision models (Jev and its alternatives) ranked by accuracy, calibration, cost per 1,000 decisions and whether they run on a CPU.

Best Decision Models, Ranked

Decision models ranked by accuracy on the Jev Decision Index: Jev and its open alternatives, with calibration, latency and cost per 1,000 decisions.

Best Calibrated Decision Models

Decision models ranked by expected calibration error (ECE): the models whose stated confidence best matches how often they are right.

Cheapest Decision Models per 1,000 Decisions

Hosted decision models ranked by cost per 1,000 decisions, excluding models below the median Decision Index score.

Best Open-Weights Decision Models

Open-weights decision models you can self-host, ranked by accuracy on the Jev Decision Index.

Best Decision Models That Run on a CPU

Small decision models that run on a CPU, ranked by accuracy on the Jev Decision Index, with latency.

Jev Alternatives: Decision Models with the Same Request Shape

Decision models that accept Jev’s request shape (state plus typed questions), ranked by accuracy on the Jev Decision Index.