No longer supported

This model is retired (released March 2023) and is no longer actively maintained or recommended by OpenAI.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

retiredlegacyllm Proprietary

Codex davinci-002

by OpenAI· Released March 2023· Cutoff June 2021

Codex davinci-002 is a deprecated version of OpenAI's Codex model, specialized in translating natural language into code. It was the most capable Codex model before being succeeded by GPT-4 and GPT-3.5 Turbo. This model is no longer available for use as of March 2023.

Official Site API Docs 📄 Research Paper

Input cost

$0.0200 per 1K tokens

Output cost

$0.0400 per 1K tokens

Context window

8000 tokens

Max output

4096 tokens

Modalities

text

License

proprietary

Capabilities

Code GenerationNatural Language to CodeCode CompletionCode ExplanationDebugging Assistance

Best For

Generating code from natural language descriptions, especially for Python and other popular languages.

Strengths

Excellent at translating natural language to code
Supports multiple programming languages
Can generate complete functions and classes
Useful for code completion and explanation

Limitations

Retired and no longer accessible
Limited to 8000 token context window
May produce insecure or inefficient code
Not as capable as newer models like GPT-4

Use Cases

Automated code generation from comments

Code completion in IDEs

Explaining code snippets

Converting pseudocode to actual code

Generating unit tests

Assisting with debugging

Learning programming by example

Back to all models