OpenAI
Lab behind the GPT series, ChatGPT and the o-series reasoning models.
Field guides
Milestones
- Sept 2024
o1 and test-time compute
A model trained with RL to reason at length before answering. More thinking at inference buys accuracy.
- May 2024
GPT-4o
One model natively handles text, audio and images, enabling real-time voice conversation.
- Feb 2024
Sora preview
Minute-long coherent video from text. Video generation joins the frontier.
- Mar 2023
GPT-4
A step change in capability on professional exams, with image input. The bar every lab now aims at.
- Nov 2022
ChatGPT
A chat interface on an RLHF-tuned model reaches a mass audience within weeks.
- Mar 2022
InstructGPT and RLHF
Reinforcement learning from human feedback makes a smaller model preferred over a far larger raw one.
- Jun 2021
GitHub Copilot preview
Code completion from a large model, inside the editor. AI coding becomes a daily tool.
- Jan 2021
CLIP
Contrastive training on image–text pairs gives a shared embedding space. It becomes the glue of multimodal systems.
- May 2020
GPT-3 and in-context learning
At 175B parameters, a model learns new tasks from a few examples in the prompt, without retraining.
- Jan 2020
Scaling laws
Loss falls as a smooth power law in parameters, data and compute, turning model building into a forecastable investment.
- Feb 2019
GPT-2 and staged release
A 1.5B-parameter model writes coherent paragraphs. OpenAI withholds full weights at first, starting the release-norms debate.