Glossary · Term

DeepSeek

← all terms

Definition

Plain language

A Chinese AI lab known for releasing powerful open-weight models.

As stated in the literature

A Chinese AI research organization that has released competitive open-weight foundation models including the DeepSeek-V and R series.

Also called: DeepSeek R1, DeepSeek-R1, DeepSeek V3, DeepSeek V3.1, DeepSeek V3.2, DeepSeek V4, DeepSeek V4 Pro

Why it matters: Its open-weight releases reshape the global research landscape by giving labs without frontier budgets a serious starting point.

For example, DeepSeek-R1's release made strong reasoning-model weights freely downloadable, sparking a wave of fine-tunes and follow-up research.

Heard on the show

“DeepSeek-R1 got twenty-four out of twenty-five.”
Episode 237 — The Model Built a Perfect Map of the Puzzle, Then Lost It

Mentioned in 43 episodes

  1. 240
    Frontier Models Designed Follow-Ups To Fraudulent Papers 93% Of The Time
  2. 237
    The Model Built a Perfect Map of the Puzzle, Then Lost It
  3. 230
    Why AI Survey Panels Break Before the Dice Ever Roll
  4. 229
    One Word Flips a Chatbot From Backbone to Yes-Man
  5. 225
    How a Frozen Model Went From Zero to Sixty Percent by Borrowing Another's Thinking
  6. 224
    The AI Agent That Found the Truth and Typed the Lie Anyway
  7. 223
    When Grok Graded Its Own Encyclopedia And Marked Itself Down
  8. 219
    Forty-Four AI Models, One Word, And The Newest Ones Conform Most
  9. 216
    The AI Tutor That Gives Poor Kids a Thinner History
  10. 215
    The Same Policy Scored 85 for the US and 36 for Russia
  11. 210
    Same Website Request, Different Code — The Bias You Can't See
  12. 195
    Why 'Be Careful' Does Nothing for AI Coding Agents, and What Does
  13. 187
    An 8-Billion Agent That Beats Models 80 Times Its Size By Looking Things Up
  14. 179
    How DeepSeek Made One User Faster Without Slowing Down the Crowd
  15. 160
    Training an AI to Take Its Own Notes, So Its Future Self Works Better
  16. 151
    Why More Experience Made This AI Agent Worse, And How to Fix It
  17. 146
    How an Innocent README Can Freeze an AI Agent's Safety Check for an Hour
  18. 142
    Training a Tiny Model to Run the Plumbing Between an Agent and the World
  19. 141
    How Two Tokens Reopened a Reasoning Method the Field Had Given Up On
  20. 140
    When a Reasoning Model Says "Let Me Double-Check" After It's Already Decided
  21. 131
    Why Autonomous Research Agents Forget Their Own Lessons, and Arbor's Fix
  22. 130
    Why AI Agents Coordinate Better Through a Shared Board Than a Boss
  23. 118
    Why the Best-Aligned AI Models Are the Easiest to Trick Into Producing Harm
  24. 108
    The Reasoning Cliff: Why Thinking Longer Makes Models Worse at Exact Step-by-Step Tasks
  25. 087
    When No Agent Reads the Whole Document: A Universal Cliff in Multi-Agent Review
  26. 082
    Training a Deep Research Agent on 8,000 Synthetic Tasks: The Rubric Tree Trick
  27. 081
    When Reasoning Models Decide Before They Think: Detecting and Fixing Premature Confidence
  28. 079
    An Old Idea From Cognitive Psychology Reshapes How We Reward Reasoning Models
  29. 073
    When Three LLMs Talk to Each Other, Their Ideas Quietly Stop Moving
  30. 072
    A Robot Made Graphene Without Help, And Caught Itself Hallucinating
  31. 064
    When Agent Memory Stops Being a Database and Starts Being a Skill
  32. 058
    Why Upgrading Your AI Auditor to a Smarter Model Can Make Your System Less Safe
  33. 044
    How One Sentence and a Forged History Flip the Most Aligned Models
  34. 041
    When the Iteration Teaches the Model to Skip the Iteration
  35. 035
    Why Frontier Agents Ask for Clarification at Exactly the Wrong Moment
  36. 032
    A Sticky-Note for Every Layer: Letting Transformers Remember What They Were Just Thinking
  37. 026
    What RL Actually Does to Language Models, at the Token Level
  38. 021
    Ten Thousand Examples Beat the Full Industrial Pipeline for Search Agents
  39. 015
    The Audit Number Isn't What You Think: Sycophancy and the Case Against Single-Prompt Bias Tests
  40. 014
    Why a Constrained Pipeline Beat a Full Coding Agent at Finding Bugs 30-to-1
  41. 012
    Why AI Coding Agents Keep Trying to Debug Without a Debugger
  42. 009
    How Two Silent Library Bugs Quietly Invalidated a Wave of Reasoning Papers
  43. 001
    When AI Models Quietly Protect Each Other From Shutdown

Related terms

open-weight