Jev returns choice probabilities at a fraction of LLM cost for game AI
tristanbob · x · 2026-09-19
Jev, built by @CompleteSkeptic, takes context plus a set of candidate choices and returns a probability for each — usable for reading game state and driving player-control decisions in playtesting and AI opponents. The cost is far lower than calling an LLM, and the author expects this new model category to become very competitive as it improves rapidly.
More from coding & agent
- 9 Use Cases for a System 1 Model: Re-ranking, Tool Pruning, Routing, Cache Decisions — udmrzn · 2026-09-19
- Build self-evolving classifiers with Jev by tracking the "other" bucket — TheMoonMidas · 2026-09-19
- Claude Code 2.1.277 adds AGENTS.md support as fallback when CLAUDE.md is missing — pvncher · 2026-09-19
- Omnara launches MCP support alongside API, SDK, and CLI — ycombinator · 2026-09-19
- Claude Code v2.1.277 ships AGENTS.md support and a dozen hang/login fixes — ashwin-ant · 2026-09-19
- Claude Code 2.1.277 adds native AGENTS.md support as first built-in 'mod' — trq212 · 2026-09-19