Models & Providers
6 min read
·┌──────────────────────────────────────────────────────────┐ │ ═══════════════════════════════════════════════════ │ │ ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ │ │ ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ │ │ ──────────────────────────────────────────────────── │ │ ██████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░ │ │ █████████████████████████████████░░░░░░░░░░░░░░░░░░ │ │ ██████████████████████████████████████░░░░░░░░░░░░░ │ │ ████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ │ │ ──────────────────────────────────────────────────── │ │ ███████████████████████████████████████░░░░░░░░░░░░ │ └──────────────────────────────────────────────────────────┘
GPT stands for "Generative Pre-trained Transformer." It's a family of language models developed by OpenAI that can understand and generate human-like text.
GPT models can:
[GPT-3]: The breakthrough model that showed what large language models could do. Released in 2020.
[GPT-3.5]: An improved version powering ChatGPT. Faster and more capable than GPT-3.
[GPT-4]: The current flagship model. Significantly more capable than GPT-3.5, better at reasoning, and can handle longer conversations.
[GPT-4 Turbo]: A faster, more efficient version of GPT-4 with updated knowledge.
GPT is a transformer model—a type of neural network architecture that's particularly good at understanding context in text. It's trained on a massive amount of text from the internet, learning patterns in how language works.
When you give GPT a prompt, it:
[Large context window]: GPT-4 can consider up to 128K tokens (roughly 100,000 words) of context.
[Multimodal capabilities]: GPT-4 Vision can understand images in addition to text.
[Fine-tuning]: You can customize GPT models for specific tasks or domains.
[Function calling]: GPT can call external tools and APIs to perform actions.
GPT has become the foundation for many AI applications and is a great starting point for understanding modern language models.