Home › Key concepts › How a language model works
Key concept · Crawl (beginner) · the same on every platform
How a language model works
It predicts the next token, again and again.
What it is
A language model is trained on huge amounts of text to predict the next token given the ones before. To write an answer it predicts one token, adds it, and predicts the next. Everything else, including reasoning and following instructions, is built on that.
Why it matters
Knowing this explains both its strengths (fluent, flexible) and its weaknesses (it can sound sure and be wrong).
An example
Given 'The capital of France is', the most likely next token is ' Paris'.
The common mistake
Treating the model as a database of facts.
Where you meet it on each platform
The idea stays the same; only the product name changes.
| Job | AWS | Azure | Google Cloud | Databricks | Open source |
|---|---|---|---|---|---|
| Model APIcall a large language model | Amazon Bedrock | Azure OpenAI in Microsoft Foundry | Gemini on Vertex AI | Foundation Model APIs | Ollama, or a free hosted open model |
Related ideas
Where it shows up
Part of the Crawl level on these paths: LLM Engineering (how models work).