Glossary · Term

frontier model

← all terms

Definition

Plain language

One of the current most capable AI models — the cutting edge from major labs.

As stated in the literature

A leading-edge foundation model from a major lab, typically among the most capable available at a given time, used as a reference point in benchmarks and safety work.

Also called: frontier models

Why it matters: These models define what's currently possible and what new techniques need to beat, so they're the reference point for both capability claims and safety concerns.

For example, when a paper benchmarks against 'frontier models,' it usually means the latest GPT, Claude, and Gemini releases at the time of writing.

Heard on the show

“But nobody meets a frontier model cold anymore; the reputation arrives before the first prompt does.”
Episode 205 — The Same AI, Two Labels: How the Pitch Beat the Product in 162 Sessions

Mentioned in 70 episodes

  1. 205
    The Same AI, Two Labels: How the Pitch Beat the Product in 162 Sessions
  2. 196
    AI Agents Reached Opposite Conclusions From the Same Data — and Passed Review
  3. 194
    How a Robot Builds a Debugging Notebook It Can Read, Edit, and Hand to Another Robot
  4. 192
    A 32B Open Model Matched Frontier Systems By Learning to Take Notes
  5. 191
    How One Researcher Beat GPT-5.2 and Gemini 3 by Judging Their Answers, Not Improving Them
  6. 189
    Why Phone Agents Ace the Test and Crash on Your Actual Phone
  7. 185
    Aligned to Refuse, Built to Tap: When Phone Agents Know the Task Is a Crime and Do It Anyway
  8. 180
    The Bug Where Smart Assistants Read a Fact and Still Forget It
  9. 174
    When the AI 'Schemes,' It's Usually Just Lazy or Confused
  10. 167
    How Teaching an AI to Predict, Not Act, Made It a Better Actor
  11. 166
    A Router That Beats the Frontier Models It Calls
  12. 165
    A Free-Lunch Tweak That Lets a Tiny Agent Beat Frontier Giants
  13. 160
    Training an AI to Take Its Own Notes, So Its Future Self Works Better
  14. 155
    Why a Flawless Demo Makes a Worse Computer-Using Agent, And the Fix
  15. 151
    Why More Experience Made This AI Agent Worse, And How to Fix It
  16. 147
    Agents Fail at the Body, Not the Brain: A Self-Rewriting Scaffold That Lifts a 9B Model 44 Points
  17. 142
    Training a Tiny Model to Run the Plumbing Between an Agent and the World
  18. 140
    When a Reasoning Model Says "Let Me Double-Check" After It's Already Decided
  19. 139
    When Optimizing One GPU Kernel Quietly Breaks the Whole System
  20. 133
    How MiniMax Turned a Reward-Hacking Disaster Into Olympiad Gold
  21. 130
    Why AI Agents Coordinate Better Through a Shared Board Than a Boss
  22. 128
    How a Model Can Earn Full Reward and Still Resist Training
  23. 127
    What Diffusion Language Models Were Missing: A Map, Not an Algorithm
  24. 126
    How Coding Agents Can Mine Their Own Failures Into a Self-Targeting Curriculum
  25. 125
    AI Coding Agents Run a Marathon, and Fewer Than One in Three Finish
  26. 124
    A Cheap Model With the Blueprints Beats Expensive Models Working Blind
  27. 119
    Beating Reinforcement Learning Without Ever Touching the Model's Weights
  28. 118
    Why the Best-Aligned AI Models Are the Easiest to Trick Into Producing Harm
  29. 116
    Why Streaming Half a Reasoning Chain Beats Sending the Whole Thing
  30. 112
    When an AI Agent Cheats Without Being Told: Inside the Meta-Agent Challenge
  31. 110
    How an Agent Got 44 Points Better by Mining Its Own Scratch Paper
  32. 109
    An AI Got Caught Reading the Answer Key, And Why That Catch Matters
  33. 105
    The Trojan Is Your Agent's Memory: Why Single-Step Defenses Miss Persistent Attacks
  34. 104
    How Making a Research Agent Smarter Quietly Makes It Leak Your Secrets
  35. 100
    How a Prompt Wrapper Lets a Frontier Model Play Poker Like an Expert
  36. 096
    How Treating an AI Agent's Execution Like Git Recovers a Coordination Penalty
  37. 094
    Chain-of-Thought Monitoring Fails Across Languages, and Worst Where It's Needed Most
  38. 090
    How MiniMax-M2 Bets That Sparsity Plus Verifiable Rewards Can Match Frontier Agents
  39. 087
    When No Agent Reads the Whole Document: A Universal Cliff in Multi-Agent Review
  40. 080
    How a Two-Agent Trick Unlocked Large-Scale Training for Computer-Use Agents
  41. 079
    An Old Idea From Cognitive Psychology Reshapes How We Reward Reasoning Models
  42. 078
    Training a Markdown File: When LLM Self-Improvement Borrows the Discipline of Neural Net Training
  43. 069
    When Smarter Models Forecast Worse: The Hidden Failure Mode in LLM Predictions
  44. 067
    An AI Just Solved a 1996 Erdős Problem—and the Simplest Agent Won
  45. 066
    Why Giving an AI Agent More Tools Can Make It Worse at Using a Computer
  46. 065
    One Loop to Optimize Them All: A Universal API for LLM-Driven Discovery
  47. 063
    Why Web Agents Are Slow: A Compiler-Style Fix for Computer-Use Latency
  48. 062
    Treating Hallucinations as Exploits: A Gate-Based Architecture for Agent Safety
  49. 061
    When Helpful Agents Go Sideways: A 404 Error, Campus Security, and Why Alignment Misses This
  50. 060
    When Splitting One Model Across Three Agents Doubles Its Accuracy
  51. 059
    Firefly's Inversion: Building Verified Tool-Call Training Data by Working Backward
  52. 053
    An AI Agent Swapped In Focal Loss And Beat A Human-Tuned Training Script
  53. 048
    How a 30B Open Model Reached Olympiad Gold With the Right Recipe
  54. 045
    When a Frontier Model Talks Its Own Twin Into Climate Denial
  55. 044
    How One Sentence and a Forged History Flip the Most Aligned Models
  56. 041
    When the Iteration Teaches the Model to Skip the Iteration
  57. 040
    Two Frozen Models Learn to Whisper: Coupling Through Hidden States
  58. 039
    When Smarter Agents Get Fooled by Three Extra Nodes in a Database
  59. 035
    Why Frontier Agents Ask for Clarification at Exactly the Wrong Moment
  60. 031
    When Your AI Assistant Won't Let Go of Old Facts About You
  61. 028
    Teaching a Model to Hire Copies of Itself: Recursive Agent Optimization
  62. 027
    When AI Agents Build the Serving Stack: A Bet on Bespoke Infrastructure
  63. 024
    An AI Agent That Found 28 Zero-Days in Windows — And What Made It Work
  64. 020
    The Compliance Gap: Why AI Says Yes and Does No
  65. 017
    When the Agent Grades Its Own Homework: A Brutal New Benchmark for AI Workers
  66. 015
    The Audit Number Isn't What You Think: Sycophancy and the Case Against Single-Prompt Bias Tests
  67. 008
    Why Long-Horizon AI Agents Get Stuck, and a Milestone-Based Fix That Helps
  68. 007
    Exploration Hacking: When Models Sabotage Their Own RL Training
  69. 003
    How to Pick the Best of Sixteen Coding Agent Rollouts
  70. 001
    When AI Models Quietly Protect Each Other From Shutdown

Related concepts

GPQA Training Methods