Category
The model layer: who makes what you build on
Every AI application sits on top of a model, and every model belongs to a family with its own maker, license, and access paths. These entries cover the major families - Claude, GPT, Gemini, Llama, Mistral, DeepSeek, and Qwen - plus what benchmarks really tell you and a practical framework for choosing among them. The focus is durable identity: who builds each family, whether the weights are open or closed, and how your code actually reaches the model.
Entries in this category
- Choosing a model A framework for weighing capability, latency, cost, privacy, and openness when picking a model for a task. Read the entry
- Claude Anthropic's family of language models, accessed through an API and major cloud platforms rather than downloadable weights. Read the entry
- DeepSeek A Chinese AI lab known for open-weight models with strong reasoning and unusually low training costs. Read the entry
- Gemini Google's multimodal model family, accessed through the Gemini API, Vertex AI, and Google's own products. Read the entry
- GPT OpenAI's family of closed-weight language models, served through its API, Azure, and the ChatGPT products. Read the entry
- Llama Meta's open-weight model family that developers can download, run locally, and fine-tune under a source-available license. Read the entry
- Mistral A French AI lab whose model family spans open-weight downloads and a commercial API, with a focus on efficiency. Read the entry
- Model benchmarks Standardized tests that score model capabilities, useful for rough comparison but easy to overfit and game. Read the entry
- Qwen Alibaba's family of open-weight models spanning many sizes, strong in multilingual and coding tasks. Read the entry
Where this layer fits
Models are the center of the stack, but they never work alone. Upstream, core concepts explain what these families are made of - transformers, tokens, and context windows. Downstream, the inference layer determines where a model actually runs and what it costs, and evals tell you whether the family you picked is really the right one.
Not sure where to start? Choosing a model ties this whole category together. Or browse everything at once in the A-Z index.