Start typing to search.
Average response time
38.47 s
Estimated 500-token answer. Across 57 models with recent complete speed data.
Average response start
5.13 s
Wait until the first token. Across 57 models with recent response start data.
Average output speed
48.2 tok/s
Generated tokens per second. Across 57 models with recent output speed data.
Prices and recent speed measurements below are for OpenAI endpoints. A 500-token response combines the wait for the first token with the time to generate 500 output tokens.
Model data fetched .
Practical comparison guide
OpenAI currently has 72 indexed models on this site, including 2 speech, 57 text, 6 image, 3 transcription, 1 video, 3 embeddings. This page combines the provider-specific catalog with published pricing and recent performance data so you can identify suitable models and then compare OpenAI with alternative hosts on each model page.
OpenAI offers model-specific routes rather than one interchangeable service. Check modality, context window, supported inputs, and parameter support before comparing price. The largest published context window in this catalog is 1.1M tokens.
The lowest published input price shown here is Free. Recent 500-token response measurements are available for 57 models; missing measurements remain visible and are never estimated.
OpenAI lists its headquarters as US. Published datacenter coverage: no datacenter regions are currently published. Headquarters and hosting regions describe different things, so verify data residency, service terms, and current availability directly with the provider.
