← Back to Models
meta-llamatransformer
Llama 3.2 90B Vision Instruct
90B parameters • 54.8GB VRAM (Q4 estimate) • 8,192 context
Estimate · Q4_K_M · 4k context · batch 1. For dense 7B–70B this sits ~5–8% above the Q4_K_M GGUF file. Not peak runtime VRAM and not a measured bench.
Specifications
Parameters
90B
VRAM (Q4 est.)
54.8 GB
VRAM (FP16)
180 GB
Context Window
8,192
Architecture
transformer
License
Apache-2.0
Finding cloud alternatives...
Run Llama 3.2 90B Vision Instruct on Radeon Instinct MI210
~54.8GB VRAM needed at Q4. Radeon Instinct MI210 has 64GB — buy hardware or rent cloud.
We may earn a commission from cloud and hardware partners at no extra cost to you.
Find the cheapest GPU that can run Llama 3.2 90B Vision Instruct
Compatibility Lab — check every GPU × quantization combination