Model reference · open weights

llama3-llava-next

llama3-llava-next is an open-weight language model from llava-hf. llama3-llava-next-8b-hf (FP16) weighs 16.7 GB; the smallest configuration that runs it is RTX 4090 24 GB.

For self-hosting, compare this build's requirements below with the GPU you plan to rent.

• llama3-llava-next is an 8.4B parameter multimodal model developed by llava-hf for image-text-to-text tasks such as image captioning and visual question answering.
• It combines a pre-trained vision encoder with the Meta-Llama-3-8B-Instruct language backbone and supports a context length of 8192 tokens.
• The model operates in English and is released under the llama3 licence.

Summary of the llava-hf/llama3-llava-next-8b-hf model card, 2026-10-01

Hugging Face ↗