Open questions on JEV: model size, calibration reliability, and a future fine-tuning API
vykthur · x · 2026-09-20
Calling JEV exciting for evals and automated agent optimization, vykthur raises three open questions: how large the model is and whether it's built on a pretrained LM base; beyond early third-party tests, how well calibrated it is and where that breaks down; and whether weak areas can be addressed — even if not now, via some future fine-tuning API.
More from coding & agent
- Jev as a Judge: cheaper, faster, more reliable evals than LLM-as-a-Judge — Hacubu · 2026-09-20
- Blender previs + Hailuo H3: a workflow for AI-rendered architectural animations — Hailuo_AI · 2026-09-20
- Parallel-sampling decision model Jev routes in ~1s vs 4-14s for regular LLMs — TigerOk4538 · 2026-09-20
- Dev builds fully autonomous trading bot in a day — it has already lost him $31,680 — ZeYanjie · 2026-09-20
- Three coding habits that may never come back in the AI-agent era — BLUECOW009 · 2026-09-20
- mcp-proxmox lets you manage Proxmox VE clusters with natural language — modelcontextprotocol · 2026-09-20