GPT (OpenAI)
OpenAI's hosted models, reached through an API. GPT is OpenAI's family of large language models, used through the OpenAI API or through Azure OpenAI. The API supports chat, structured output, tool calling, embeddings, vision and speech. Teams use it to build assistants, agents, document processing and internal automation without running any model infrastructure.
When it is the right choice
- You want strong general-purpose models without operating GPUs
- You need features such as tool calling, structured output and multimodal input in one API
- You need enterprise contracts through Azure OpenAI in a specific region
- Not ideal when data cannot leave your own infrastructure; self-hosted open-weight models fit better there
How we use it
- Put an API gateway in front of the model for keys, rate limits, retries and cost tracking
- Trace every call with OpenTelemetry so cost, latency and failures are visible per feature
- Build evaluation sets so model upgrades are tested rather than assumed
- Design fallbacks to a second provider for critical paths
Azure
LangChain
pgvector
OpenTelemetryWhere it fits in our work
Related articles
From Idea to AI Feature in Six WeeksHow an AI project runs end to end in six weeks: discovery, design, prototype, infrastructure, rollout, then measuring and improving.
CI/CD for Models and PromptsEvaluation gates, model registries, canary releases and rollback: applying CI/CD discipline to prompts and models in LLM applications.
Fine-Tuning or Prompting? A Decision Guide for Small TeamsWhen prompting, RAG or fine-tuning is the right way to adapt a language model, and what each one costs a small engineering team.GPT (OpenAI): frequently asked
Does OpenAI train on data sent through the API?
OpenAI states that API data is not used for training by default, but read the current data usage policy and your contract terms. Azure OpenAI has its own data handling terms.
How do we keep API costs under control?
Track token usage per feature, cache repeated requests, use smaller models where they pass your evaluations and set budget alerts. We build these controls into the gateway from day one.
Running GPT (OpenAI), or deciding whether to? Tell us what you have; the first consultation is free.
Let's talk »
