LLaVA models

9 LLaVA models from llava-hf on Hugging Face, from 894M to 13.4B parameters, published by llava-hf in F16 and BF16, with contexts from 4K to 32K tokens. The smallest official model, llava-onevision-qwen2-0.5b-ov-hf, needs about 2.0 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.088/hr.

Pick a size

One row per official LLaVA 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
llava-onevision-qwen2-0.5b-ov-hf894M2.0 GB1.0 GB0.5 GBRTX 3070$0.088/hr0.38 GB
llava-1.5-7b-hf7.1B15.8 GB7.9 GB3.9 GBV100$0.110/hrnot published for this architecture
llava-v1.6-mistral-7b-hf7.6B16.9 GB8.5 GB4.2 GBRTX A5000$0.176/hrnot published for this architecture
llava-1.5-13b-hf13.4B29.8 GB14.9 GB7.5 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 (6)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
llava-onevision-qwen2-0.5b-ov-hf894MF162.0 GBRTX 3070$0.088/hr
llava-v1.6-vicuna-7b7.1BBF1615.8 GBRTX 5060 Ti$0.145/hr
llava-1.5-7b-hf7.1BF1615.8 GBV100$0.110/hr
llava-v1.6-mistral-7b7.6BBF1616.9 GBRTX A5000$0.176/hr
llava-v1.6-mistral-7b-hf7.6BF1616.9 GBRTX A5000$0.176/hr
llava-1.5-13b-hf13.4BF1629.8 GBV100$0.220/hr

Fine-tunes and community models (3)

ModelParametersPublished asVRAM neededLive GPU fitEst. $/hr
llava-med-v1.5-mistral-7b7.6BBF1616.9 GBRTX A5000$0.176/hr
llava-onevision-qwen2-7b-ov8.0BBF1617.9 GBRTX A5000$0.176/hr
LLaVA-Video-7B-Qwen28.0BBF1617.9 GBRTX A5000$0.176/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.

Start ComfyUI on a rented GPU.

© 2026 Aquanode. All rights reserved.

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