No longer supported

This model is legacy (released March 2023) and is no longer actively maintained or recommended by OpenAI. Consider using their current flagship model instead.

legacyfrontierllm Proprietary

GPT-4-32k

by OpenAI· Released March 2023· Cutoff September 2021

GPT-4-32k is a variant of OpenAI's GPT-4 model with an extended context window of 32,768 tokens, enabling processing of larger documents and conversations. It offers the same advanced reasoning, creativity, and safety as GPT-4 but with a higher context limit. This model is now legacy and replaced by newer models like GPT-4 Turbo.

Official Site API Docs 📄 Research Paper

Input cost

$60.00 per 1M tokens

Output cost

$120.00 per 1M tokens

Context window

32K tokens

Max output

8192 tokens

Modalities

text

License

proprietary

Capabilities

Function CallingCode GenerationStreamingJSON ModeReasoningCreative WritingSummarizationTranslation

Best For

Tasks requiring processing of very long documents or conversations with high-quality reasoning.

Strengths

Extended context window of 32K tokens
Strong reasoning and problem-solving abilities
High-quality creative and technical writing
Supports function calling for tool integration

Limitations

Legacy model, no longer recommended for new projects
Higher cost compared to newer models
Slower inference than GPT-4 Turbo
No vision or multimodal capabilities

Use Cases

Analyzing long legal or financial documents

Processing entire books or research papers

Long-form content generation

Complex multi-turn conversations with extensive history

Code analysis of large codebases

Summarizing lengthy reports

Data extraction from large texts

Improvements Over Previous Model

Extended context window from 8K to 32K tokens compared to GPT-4-8k
Same core capabilities as GPT-4 but with larger context
Improved ability to handle long documents and conversations
No other improvements over GPT-4-8k; this is a context size variant

Back to all models