Text Generation ModelDeprecated

GPT-3

GPT-3 (text-davinci-003) is an instruct-tuned language model from OpenAI with a 4,096-token context window.

Start Building with GPT-3 View All Models

Publisher

OpenAI

TypeInstruct

Context Window4,096 tokens

Replaced byGPT-5 nano

Try GPT-3 →

About GPT-3

Instruct-tuned language generation from OpenAI

Text-davinci-003 is an instruct-tuned variant in OpenAI's GPT-3.5 model series, designed to follow natural language instructions and produce detailed, contextually relevant text. It was made available through the OpenAI API and is optimized for tasks that require nuanced language understanding, including complex question answering, creative writing, and summarization. The model accepts text input and generates text output, with a maximum context window of 4,096 tokens and a maximum response size of 2,500 tokens.

Text-davinci-003 is particularly suited for applications that involve generating long-form content, drafting structured responses, or handling multi-step language tasks within a single prompt. Compared to earlier GPT-3 variants, it incorporates instruction-following improvements that make it more reliable for task-specific prompting. OpenAI has since deprecated this model in favor of newer offerings, so it is no longer recommended for new production integrations.

Capabilities

What GPT-3 supports

Instruction Following

Responds to natural language instructions to complete tasks, a core feature of the instruct-tuned training approach used for this model.

Long-Form Text Generation

Generates extended prose, articles, or structured documents up to a maximum response size of 2,500 tokens per completion.

Summarization

Condenses longer input text into concise summaries, operating within the model's 4,096-token context window.

Question Answering

Answers factual and open-ended questions based on context provided in the prompt, leveraging the model's language understanding capabilities.

Creative Writing

Generates creative content such as stories, dialogue, and poetry based on user-supplied prompts and constraints.

Text Classification

Categorizes or labels input text according to criteria specified in the prompt, without requiring fine-tuning.

Ready to build with GPT-3?

Get Started Free

FAQ

Common questions about GPT-3

What is the context window size for text-davinci-003?

Text-davinci-003 has a context window of 4,096 tokens, which includes both the input prompt and the generated response.

What is the maximum response length this model can produce?

The model supports a maximum response size of 2,500 tokens per completion.

Is text-davinci-003 still available for use?

Text-davinci-003 has been officially deprecated by OpenAI. It is no longer recommended for new integrations, and OpenAI has directed developers to use newer models in its place.

Does text-davinci-003 support image or video inputs?

No. Text-davinci-003 is a text-only model and does not support image or video analysis.

What type of model is text-davinci-003?

It is an instruct-tuned large language model, meaning it was trained to follow natural language instructions rather than simply completing text in a generative fashion.

Resources