Activelegacyllm Open Source

Code Llama 13B

by Meta· Released August 2023· Cutoff January 2023

Code Llama 13B is a large language model specialized for code generation and understanding, based on Llama 2. It is part of Meta's Code Llama family, offering a balance between performance and efficiency for code-related tasks. The model excels in generating code from natural language prompts and supports multiple programming languages.

Official Site 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

16K tokens

Max output

—

Modalities

text

Parameters

13B

License

Llama 2 Community License

Capabilities

Code GenerationCode CompletionCode InfillingCode ExplanationNatural Language to CodeMulti-language Support

Best For

Code generation and understanding tasks where a balance of performance and resource efficiency is needed.

Strengths

Strong code generation capabilities across multiple languages
Open source and freely available
Good balance of size and performance for code tasks

Limitations

Smaller context window compared to newer models
Not as powerful as larger Code Llama variants (34B)
Limited to text modality, no vision or audio support

Use Cases

Generating code from natural language descriptions

Completing partial code snippets

Explaining code functionality

Debugging and fixing code errors

Translating code between programming languages

Generating unit tests

Code review assistance

Improvements Over Previous Model

Introduced specialized code generation capabilities over Llama 2
Supports code infilling (filling in missing code) in addition to generation
Trained on a larger corpus of code data compared to Llama 2
Achieves state-of-the-art performance on code benchmarks at time of release

Back to all models

Activelegacyllm Open Source

Code Llama 13B

by Meta· Released August 2023· Cutoff January 2023

Official Site 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

16K tokens

Max output

—

Modalities

text

Parameters

13B

License

Llama 2 Community License

Capabilities

Code GenerationCode CompletionCode InfillingCode ExplanationNatural Language to CodeMulti-language Support

Best For

Code generation and understanding tasks where a balance of performance and resource efficiency is needed.

Strengths

Strong code generation capabilities across multiple languages
Open source and freely available
Good balance of size and performance for code tasks

Limitations

Smaller context window compared to newer models
Not as powerful as larger Code Llama variants (34B)
Limited to text modality, no vision or audio support

Use Cases

Generating code from natural language descriptions

Completing partial code snippets

Explaining code functionality

Debugging and fixing code errors

Translating code between programming languages

Generating unit tests

Code review assistance

Improvements Over Previous Model

Introduced specialized code generation capabilities over Llama 2
Supports code infilling (filling in missing code) in addition to generation
Trained on a larger corpus of code data compared to Llama 2
Achieves state-of-the-art performance on code benchmarks at time of release

Back to all models