No longer supported

This model is legacy (released July 2023) and is no longer actively maintained or recommended by Meta.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

legacylegacyllm Open Source

Llama 2 13B

by Meta· Released July 2023· Cutoff September 2022

Llama 2 13B is a 13-billion-parameter large language model from Meta, part of the Llama 2 family released in July 2023. It is a successor to Llama 1, trained on 2 trillion tokens and optimized for dialogue use cases with fine-tuning on human preferences. The model is open-source and available for both research and commercial use.

Official Site 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

4096 tokens

Max output

4096 tokens

Modalities

text

Parameters

13B

License

Llama 2 Community License

Capabilities

Text GenerationCode GenerationDialogueInstruction FollowingZero-shot LearningFew-shot Learning

Best For

General-purpose text generation and dialogue applications with a balance of performance and efficiency.

Strengths

Strong performance on reasoning and coding tasks for its size
Open-source with permissive license
Fine-tuned for helpfulness and safety
Good balance between speed and quality

Limitations

Smaller context window (4096 tokens) compared to newer models
No native function calling or structured output support
Outperformed by Llama 3 and other newer models
Limited multilingual capabilities

Use Cases

Chatbots and virtual assistants

Text summarization

Code generation and explanation

Content creation and drafting

Educational tutoring

Research and experimentation

Fine-tuning for domain-specific tasks

Improvements Over Previous Model

Trained on 40% more data (2 trillion tokens vs 1 trillion)
Double the context length (4096 vs 2048 tokens)
Fine-tuned with RLHF for better safety and helpfulness
Open-source with commercial use allowed (unlike Llama 1)
Improved performance on reasoning, coding, and knowledge benchmarks

Back to all models

No longer supported

This model is legacy (released July 2023) and is no longer actively maintained or recommended by Meta.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

legacylegacyllm Open Source

Llama 2 13B

by Meta· Released July 2023· Cutoff September 2022

Official Site 🤗 Hugging Face 📄 Research Paper

Input cost

Free (open source)

Output cost

Free (open source)

Context window

4096 tokens

Max output

4096 tokens

Modalities

text

Parameters

13B

License

Llama 2 Community License

Capabilities

Text GenerationCode GenerationDialogueInstruction FollowingZero-shot LearningFew-shot Learning

Best For

General-purpose text generation and dialogue applications with a balance of performance and efficiency.

Strengths

Strong performance on reasoning and coding tasks for its size
Open-source with permissive license
Fine-tuned for helpfulness and safety
Good balance between speed and quality

Limitations

Smaller context window (4096 tokens) compared to newer models
No native function calling or structured output support
Outperformed by Llama 3 and other newer models
Limited multilingual capabilities

Use Cases

Chatbots and virtual assistants

Text summarization

Code generation and explanation

Content creation and drafting

Educational tutoring

Research and experimentation

Fine-tuning for domain-specific tasks

Improvements Over Previous Model

Trained on 40% more data (2 trillion tokens vs 1 trillion)
Double the context length (4096 vs 2048 tokens)
Fine-tuned with RLHF for better safety and helpfulness
Open-source with commercial use allowed (unlike Llama 1)
Improved performance on reasoning, coding, and knowledge benchmarks

Back to all models