← Back to docs

Which model should I use?

Match your task to the right model. Start with the use case, then pick the recommended option.

Chatgpt-4o

Fast, multimodal, great for conversation and general tasks.

Reasoninggpt-5-codex

Deep logic, math, and multi-step problem solving.

Cheap / high-volumeLlama-4-Maverick

Open-weight, cost-effective for bulk inference.

Visiongpt-4o or claude-opus-4-8

Image understanding, OCR, and visual Q&A.

Speechwhisper-1 / tts-1

Transcription and text-to-speech pipelines.

Still unsure? Start with gpt-4o — it handles most tasks well. Switch to a specialized model when you hit a ceiling.