CONTROLAgents & automation
Jev: A Decision-Focused AI Model
AI Adventurer@AiAdventurerx
2:56Jev is an AI model developed by TypeSafe AI that is trained for making decisions using Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning from AI Feedback (RLAIF) or similar techniques, rather than generating text. It is designed for high-frequency, low-cost decision-making in agentic systems.
ChatGPT’s RLHF co-creator just shipped a model that can’t write.
Jev (TypeSafe AI) is trained with RLCD for decisions, not paragraphs. They claim 20–200x faster and 40–400x cheaper, with output tokens free. Input is $0.042 / MTok.
The trade: no text generation. Code + structured choices. Wikipedia-race and real-time loops are the demos.
If agents are bottlenecked on cheap, high-frequency decisions, this is the architecture bet.
Blog: typesafe.ai/blog/introducing-system-one-mo
Waitlist: typesafe.ai
#jev #typesafe #llm #aiagents #rlhf
x.com/CompleteSkeptic/status/2099925682726