AI Model Timeline

Tracking the accelerating release frequency of frontier AI models.

GPT-2

OpenAI·

LLMsflagshipcloud

Overview

Language model that was initially deemed too dangerous to release.

Capabilities and innovations

1.5B ParametersCoherent text generationZero-shot learningScaled Transformer ArchitectureZero-shot Task TransferWebText Dataset

Benchmarks

No benchmark values are recorded for this model.

  • GPQA Diamond: benchmark did not exist at release
  • Humanity’s Last Exam (no tools): benchmark did not exist at release
  • SWE-bench Verified: benchmark did not exist at release
  • SWE-bench Pro: benchmark did not exist at release
  • MMLU: benchmark did not exist at release
  • MMLU-Pro: benchmark did not exist at release
  • Terminal-Bench 2.x: benchmark did not exist at release
  • MMMU-Pro: benchmark did not exist at release

Architecture and hardware

Parameters
1.5B total (Dense)

Links

More from OpenAI

Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.