Choosing between hosted AI models
6 ordered steps, from OpenAI to DeepSeek.
AI providers and models
-
Step 1: OpenAI
Start with a familiar hosted provider so the learner can anchor the rest of the model vocabulary to a real API ecosystem.
-
Step 2: GPT (model family)
After the provider, the model family explains what kind of engine the application is actually selecting.
-
Step 3: Chat model
Chat models are the baseline interaction pattern, so they set expectations for ordinary assistant-style request and response work.
-
Step 4: Reasoning model
Reasoning models matter after chat models because they change the tradeoff from fast conversation toward deeper multi-step problem solving.
-
Step 5: DeepSeek-R1
DeepSeek-R1 makes the reasoning-model idea concrete with an example learners may encounter when comparing cost and capability.
-
Step 6: DeepSeek
Ending on the provider helps separate a specific reasoning model from the broader organization and model ecosystem behind it.