TypeSafe AI launched Jev on 15 September 2026: a model that returns typed values with calibrated probabilities instead of text, at $0.042 per million input tokens with output billed at nothing. Its workflow evals use the average of GPT-6 Astra and Claude Fable 5.1 as the reference answer, so scores measure agreement rather than correctness, and on TypeSafe’s own chart Jev ties Sonnet 5 at 67.8 per cent. Three independent tests measured roughly 5x faster and 8.6x cheaper than Mistral Small 4, and 1.6x cheaper than DeepSeek V4.1 Flash.
