Skip to content
← Models
ClosedOpenAI/10 trustVerified
GPT-4 Turbo logo

GPT-4 Turbo

128K-context GPT-4 upgrade announced at OpenAI DevDay 2023.

Visit official page →

About

GPT-4 Turbo was unveiled on November 6, 2023, at OpenAI's first developer conference, DevDay, as a faster and cheaper successor to GPT-4. It expanded the context window to 128,000 tokens, updated world knowledge to April 2023, and added developer controls including JSON mode, parallel function calling, reproducible outputs via seed parameters, and log probabilities.

It became the default GPT-4-class API model through much of 2024, available through OpenAI, Azure OpenAI, and ChatGPT, with vision input (GPT-4 Turbo with Vision) added for image understanding. It was later superseded by GPT-4o as the recommended option, but remains a landmark in the shift from 8K to 100K+ context windows. Access is API-only; weights were never released.

Specs

Context window

Pricing

Free quota

API access

Weights

Closed — API only

Released

November 6, 2023

Benchmarks

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

long-context-analysisapi-developmentdocument-qa

Trust score — 9/10

/10 trustVerified

Scored 9/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 75; Trending score 88/100 from researcher assessment.

More in the gpt family

Sources