Llama 2 13B HF vs Llama 2 70B HF

Side-by-side comparison of VRAM requirements, quantization, context length, and hardware compatibility.

Llama 2 13B HF

Meta · 13.0B

Chat
Llama 2 70B HF

Meta · 69.0B

Chat

Specifications

Llama 2 13B HFLlama 2 70B HF
Parameters13.0B69.0B
Context
Architecture
LicenseLlama 2 CommunityLlama 2 Community
Downloads35.0K15.4K
ReleasedApr 2024

VRAM by Quantization: Llama 2 13B HF vs Llama 2 70B HF

QuantizationBitsLlama 2 13B HF VRAMLlama 2 70B HF VRAM
BF1616.0028.6 GB151.8 GB

Verdict

Llama 2 13B HF needs less VRAM at BF16 (28.6 GB vs 151.8 GB), so it fits on smaller GPUs. Llama 2 13B HF is the more widely downloaded of the two.

Frequently Asked Questions

Which needs less VRAM, Llama 2 13B HF or Llama 2 70B HF?

At BF16, Llama 2 13B HF needs 28.6 GB and Llama 2 70B HF needs 151.8 GB, so Llama 2 13B HF is the lighter option to run locally.

What is the difference between Llama 2 13B HF and Llama 2 70B HF?

Llama 2 13B HF is a 13.0B model from Meta (Llama 2 family), while Llama 2 70B HF is a 69.0B model from Meta (Llama 2 family). Compare their VRAM requirements above to see which fits your GPU or Mac.