Model Intelligence

GPT-4 Turbo

Frontier Languageby OpenAIReleased November 2023

GPT-4 Turbo extended GPT-4's context window to 128K tokens, reduced API pricing by 3x, and added knowledge up to April 2024. It remained OpenAI's workhorse production model until GPT-4o superseded it on capability-per-cost metrics in mid-2024.

Context Window
128,000 tokens
License
Proprietary
Capabilities
  • 128K context window
  • Strong reasoning and coding
  • Function calling and JSON mode
  • Vision capabilities (image understanding)
  • Knowledge through April 2024
Pricing Summary
Input: $10.00/1M tokens | Output: $30.00/1M tokens. Largely superseded by GPT-4o in cost efficiency.
Benchmarks
MMLU: 86.4% | HumanEval: 87.1% | MATH: 72.2% | GSM8K: 93.7%
Competitive Positioning
Being deprecated in favor of GPT-4o in most use cases due to GPT-4o's lower cost and native multimodal capability.
Related Models
GPT-4oGPT-3.5 Turboo1