Jev beats Gemini 2.5 Flash Lite on classifier eval quality and speed
Malte Ubl ran TypeSafe AI's Jev against an existing classifier eval, where it saturated the eval on quality and was 6x faster than Gemini 2.5 Flash Lite.
Simulating Product Adoption Intent of 150 Fictional Personas via Jev API
The author built a system that sends 12 questions via the Jev API to each of 150 fictional personas to ask about their product adoption intent. The actual cost was 1.8 yen and it took about 5 seconds—fast enough, though latency is included.
I built a system to ask 150 fictional personas on the trending Jev about their product adoption intent. It throws 12 questions via API for each of the 150 personas. The actual cost was 1.8 yen, and the time taken was about 5 seconds. It's plenty fast, but it includes latency
· 1.1K likesOpen on X
More like this
Fast row classification in DuckDB with Jev
The author released a DuckDB extension that uses TypeSafe AI's Jev to quickly classify rows in CSV/Parquet files or DuckDB tables, taking about 10 seconds for 1k rows—faster than an LLM and more ergonomic than a classifier.
Classifying 1,018 AI Papers with Jev for $0.08
The author summarized each paper with DeepSeek V4 Flash, then sent the title, summary, and 24 candidate topics to Jev for classification, processing 1,018 AI papers at $0.08 total cost and 256ms median end-to-end latency per paper.
