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.
Third-Party Test Compares FLock Models and Jev System One Accuracy
A third party tested accuracy on 68 questions: FLock's THIS/THAT Model 1.0 scored 94.1% and hosted System One service Jev scored 76.5%; the post notes encouraging results for a specialized decision model, while complex tasks like graph traversal and multi-step arithmetic remain limited.
제3자가 정확도 검증을 위해 기록한 68개 질문 테스트 결과: FLock의 THIS / THAT Model 1.0: 94.1% Jev (호스팅형 System One 서비스): 76.5% 특화된 의사결정 모델로서는 고무적인 결과입니다. 다만 그래프 탐색이나 여러 단계의 산술 연산이 필요한 복잡한 작업은 여전히 한계로 남아 있습니다.
· 0 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.
