New Local-LLM Primitive: "Span" Answers Point to Snippets Instead of Enumerating
generativist · x · 2026-09-22
Developer mmastrac introduces a new "Jev-at-home" primitive: span answers, where the model indicates which snippets best answer a question instead of enumerating answers — useful when the question already contains its candidates. He says you'll be able to try it soon in DiffusionGemma-as-Jev (aka djev), and a booster predicts Jev itself will likely ship this feature.
More from coding & agent
- Simonw on 'decision models': evals matter even more than for regular LLM projects — HamelHusain · 2026-09-22
- Andrew Ng's team open-sources OpenWorker, a local desktop AI coworker (18k stars) — tom_doerr · 2026-09-22
- Dev pitches screenless voice-first workflow: pendant to local transcription to agent router — EricBuess · 2026-09-22
- Open-source Office engine Univer ships Office Harness, giving AI agents a shared workspace runtime — Aiden_Tech_Ai · 2026-09-22
- Jev Workflows 0.3 RC ships as open-source Codex plugin for decisions, failure diagnosis and completion checks — BLUECOW009 · 2026-09-22
- mcp-server-npm-plus: An MCP Server for Searching npm Packages and Scanning Vulnerabilities — modelcontextprotocol · 2026-09-22