Developer floats using jev-like models to fill a frontier model's custom 256-parameter embedding space
tobowers · x · 2026-10-13
Developer tobowers asks whether anyone has used a jev-like model to populate a custom embedding space created by a frontier model. The idea: have a frontier model analyze your data and build a compact 256-parameter embedding space, then let the small jev-like model embed any new incoming content into that space. The post is an open question with no implementation or benchmarks yet.
More from Models
- Theo benchmarks Opus 5.5, GPT-6.1 Sol and Muse Wonder Full on a bug fix — shadcn · 2026-10-13
- Claude Code at $100/month plus $20 ChatGPT: AI subscription costs stack up — dumpshoot · 2026-10-13
- Grok subscription is currently offering a 50% discount — sujingshen · 2026-10-13
- Pleias team on leaderboards: build internal evals for real use cases, but can't ignore them — antoine_chaffin · 2026-10-13
- New Paper Explains Mamba-3's State Rotations as Underdamped Second-Order Dynamics — tri_dao · 2026-10-13
- StepFun teases open-source Step 5 Preview, a flagship model for agentic tasks — bdsqlsz · 2026-10-13