Which model, right now
One answer per job — the single model to reach for today, drawn from live quality, speed, and cost data. No filters, no comparison grid.
One answer per job — the single model to reach for today, drawn from live quality, speed, and cost data. No filters, no comparison grid.
Updated 17h ago
Long agent loops, tool calling, and hard multi-step reasoning.
Writing and reviewing code, resolving real issues.
Compares API pricing only — subscription tools like Claude Code and Codex sit outside this ranking and are often the strongest option if you're already paying for a seat.
Prose, docs, and anything a human will actually read.
Reading images, screenshots, and documents.
High-volume tasks where cost per token dominates.
Chat UIs and voice, where time to first token matters.
Whole codebases and huge documents in one prompt.