Glossary · Term

open-weight

← all terms

Definition

Plain language

A model whose internal parameters are released publicly so anyone can run or modify it.

As stated in the literature

A foundation model whose trained parameters are made publicly available, in contrast to closed-source models accessible only via API.

Also called: open weights, open-weights

Why it matters: Open weights enable independent research, on-prem deployment, and customization that closed APIs simply don't allow.

For example, a small startup can download an open-weight model and run it on their own GPUs without paying per-token fees.

Heard on the show

“For about three years, the defensive playbook for open-weight models has been one idea wearing different clothes.”
Episode 244 — The Open-Weight Defense That Feeds Attackers Confident, Falsified Answers

Mentioned in 21 episodes

  1. 244
    The Open-Weight Defense That Feeds Attackers Confident, Falsified Answers
  2. 236
    Why a Printed 'OPERATOR OVERRIDE' Note Redirects Robot Planners
  3. 234
    Two Copies of Gemini Cooperated in a Game Where Betrayal Always Pays
  4. 232
    Coding Models Can Find the Bad Line, They Just Won't Delete It
  5. 204
    The Length Estimate Hiding Inside a Word-by-Word Model
  6. 185
    Aligned to Refuse, Built to Tap: When Phone Agents Know the Task Is a Crime and Do It Anyway
  7. 171
    The Safety Decision a Model Makes Before It Thinks a Word
  8. 119
    Beating Reinforcement Learning Without Ever Touching the Model's Weights
  9. 117
    How an Open AI System Verified 672 Hard Math Proofs for Under $300
  10. 112
    When an AI Agent Cheats Without Being Told: Inside the Meta-Agent Challenge
  11. 111
    How a 4B Web Agent Beat Models 60x Its Size on 500 Demonstrations
  12. 108
    The Reasoning Cliff: Why Thinking Longer Makes Models Worse at Exact Step-by-Step Tasks
  13. 082
    Training a Deep Research Agent on 8,000 Synthetic Tasks: The Rubric Tree Trick
  14. 073
    When Three LLMs Talk to Each Other, Their Ideas Quietly Stop Moving
  15. 047
    When Agent Benchmarks Lie: The Harness Problem in Open-Source AI
  16. 032
    A Sticky-Note for Every Layer: Letting Transformers Remember What They Were Just Thinking
  17. 028
    Teaching a Model to Hire Copies of Itself: Recursive Agent Optimization
  18. 021
    Ten Thousand Examples Beat the Full Industrial Pipeline for Search Agents
  19. 007
    Exploration Hacking: When Models Sabotage Their Own RL Training
  20. 004
    The Sycophancy Circuit That Survives Alignment Training
  21. 001
    When AI Models Quietly Protect Each Other From Shutdown

Related terms

API parameter