Tag
Nagi is a new open-source fast decision model that consistently outperforms other models like laya and semif in real-time gaming scenarios.
FWBench introduces a benchmark for evaluating how language models select and use time-series forecasts to make cost-constrained decisions, comparing hosted and local configurations on electricity and cycle-hire datasets with efficient budget usage by GPT-6 Astra.
This article provides a step-by-step guide to building an agentic harness using Jev, including code, architecture, and decision-making processes for AI development.
The author has open-sourced Jev Decisions v1, a dataset of 12 million examples for training AI models on agentic decisions like tool selection and routing, to address data gaps in agent decision-making.
Shane Parrish interviews Shopify CEO Tobi Lütke on topics including AI usage at Shopify, osmosis learning, and using AI for strategic decision making.
Stuntd is an open-source local proxy that records LLM decision calls, trains a lightweight model to handle them locally, reducing costs while maintaining high agreement with the teacher model.
Jev-Omni is the first open-weight model to extend typed decisions to multimodal, supporting text, images, audio, and video, and directly returning the probability distribution of options without generating explanations.
Tavily integrates with Jev to provide real-time web intelligence for research agents, enabling efficient decision-making and reducing LLM inference overhead.
Jev is a tool that acts as a layer between deterministic software and LLMs, optimized for semantic judgment tasks where possible answers are predefined but input interpretation requires fuzzy logic, such as in support-ticket routing.
The article discusses the ethical boundaries of AI agent autonomy, questioning which actions should require human approval even if technically feasible, using refund decisions as an example.
Jev is a new AI model by TypeSafe AI optimized for rapid, low-cost decision-making, offering significant speed and cost improvements over traditional AI models.
JEV, an AI model for rapid decision-making, is now integrated into Spawn, enabling game creators to design dynamic NPCs, bosses, and puzzles that respond intelligently.
The author created a cat survival game to test the performance of AI models Jev and Laya, finding that Jev is more accurate but slower, while Laya is fast but inaccurate, and plans to develop a leaderboard for such models.
The post compares AI models Jev and Laya, highlighting Laya's open-source, local, and low-latency capabilities for real-time decision-making in games.
This paper introduces Taste-Bench, a benchmark for measuring taste in LLM agents' long-horizon decisions, finding that frontier models have low accuracy and that taste can be improved through distillation training.
Kev is a family of small decision models built on Qwen3.5, offering pretrained weights and training code for yes/no, multiple-choice, and rating questions. It includes a web playground and is compatible with TypeSafe's System One API.
Shane Parrish discusses with Shopify CEO Tobi Lütke how intuition becomes crucial for decision-making in the absence of immediate feedback, emphasizing the importance of long-term strategic investments over short-term gains.
A user describes using Jev to analyze 468 tweets and identify the strongest hooks without view counts, enabling 2,340 decisions in 60 seconds for 1.3 cents, showcasing AI's efficiency in decision-making.
JevHarness is an open-source tool that allows LLMs to automatically generate and optimize task-specific decision flows for Jev, enabling self-optimization and improving performance, as shown in a Pokémon AI battle win rate increase from 25% to 75%.
A ChatGPT co-inventor has launched a new AI focused on decision-making, with speeds 20-200 times faster, already used in scenarios like Doom, Minecraft, and browser agents.