Llama 2 models

8 Llama 2 models from Meta on Hugging Face, from 6.7B to 13.0B parameters, published by Meta in F16, with a 4K-token context. The smallest official model, Llama-2-7b-chat-hf, needs about 15.1 GB of VRAM at its published precision; the cheapest live fit is V100 at $0.110/hr.

Part of the Llama series · Previous generation: Llama 1 · Next generation: Llama 3

Pick a size

One row per official Llama 2 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.

ModelParametersNative VRAMFP8 VRAMINT4 VRAMLive GPU fit (native)Est. $/hrKV cache at 32K
Llama-2-7b-hf6.7B15.1 GB7.5 GB3.8 GBV100$0.110/hr16.0 GB
Llama-2-13b-chat-hf13.0B29.1 GB14.5 GB7.3 GBV100$0.220/hr25.0 GB

VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.

Official models (4)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
Llama-2-7b-chat-hf6.7BF1615.1 GBV100$0.110/hr
Llama-2-7b-hf6.7BF1615.1 GBV100$0.110/hr
Llama-2-13b-chat-hf13.0BF1629.1 GBV100$0.220/hr
Llama-2-13b-hf13.0BF1629.1 GBV100$0.220/hr

Fine-tunes and community models (4)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
llama-2-7b6.7BF1615.1 GBV100$0.110/hr
llama-2-7b-chat6.7BF1615.1 GBV100$0.110/hr
HarmBench-Llama-2-13b-cls13.0BBF1629.1 GBRTX A6000$0.363/hr
llama2-13b-chat-hf13.0BF1629.1 GBV100$0.220/hr

VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. See the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.

Run and fine-tune Llama 2

More from Meta

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.