Developer Taro L. Saito (@taroleo) followed a TypeSafe Jev-style approach—dropping full text/JSON generation and specializing on decisions—to distill DeepSeek V4 Flash judgments (~157GB weights) into a 4B model with about 26 hours of training on a DGX Spark. The student is roughly 1/20 the size, beats the teacher’s instant-answer mode, and runs about 22ms per decision, showing a fast local-LLM decision engine is practical.

Key Takeaways

  • ✓Jev-style decision-only specialization distills DeepSeek V4 Flash judgments into a 4B model.
  • ✓About 26 hours on a DGX Spark; roughly 1/20 the teacher size and ~22ms per decision.
  • ✓Claims to beat the teacher's instant-answer mode—useful for local routing/gating agents.
⚡

Key Decision Metrics at a Glance

TL;DR Verdict
🏆 Benchmark RankRank
See how it ranks on the leaderboard
Full Leaderboard→
💰 Cost & PlansCompare
DeepSeek-V4-Flash · in: $0.22 / out: $0.66
Compare Plans↗
🛠️ Ecosystem & IDERules
DeepSeek V4 Flash · distillation · Jev
Setup IDE Rules↗
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

⚔️ Compare Model→💰 Compare plans & limits↗
Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

⚔️ Launch VS Compare→💳 View Plans↗
Up NextvLLM Semantic Router open-sources Decision 3.0: six multimodal decision models from 0.6B to 27B, #1 on Jev Decision Index in its own evaluation
Read Now→