Llama 3.2 90B Vision Instruct — Hardware Requirements & GPU Compatibility
Vision
Llama 3.2 90B Vision Instruct is a 88.6B-parameter open language model from Meta in the Llama 3 family. At BF16 it needs about 194.91 GB of VRAM — see which GPUs and Macs can run it below.
39.0K downloads
359 likesSpecifications
- Publisher
- Meta
- Family
- Llama 3
- Parameters
- 88.6B
- Release Date
- 2024-09-19
- License
- llama3.2
Get Started
HuggingFace
How Much VRAM Does Llama 3.2 90B Vision Instruct Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16est. | 16.00 | 194.9 GB | — | 177.19 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Llama 3.2 90B Vision Instruct?
BF16 · 194.9 GBLlama 3.2 90B Vision Instruct (BF16) requires 194.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 254+ GB is recommended. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Llama 3.2 90B Vision Instruct?
BF16 · 194.9 GB3 devices with unified memory can run Llama 3.2 90B Vision Instruct, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomBenchmarks
Benchmark details →Related Models
Frequently Asked Questions
- How much VRAM does Llama 3.2 90B Vision Instruct need?
Llama 3.2 90B Vision Instruct requires 194.9 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 88.6B × 16 bits ÷ 8 = 177.2 GB
KV Cache + Overhead ≈ 17.7 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
BF16194.9 GB- Can NVIDIA GeForce RTX 5090 run Llama 3.2 90B Vision Instruct?
No — Llama 3.2 90B Vision Instruct requires at least 194.9 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Llama 3.2 90B Vision Instruct on a Mac?
Llama 3.2 90B Vision Instruct requires at least 194.9 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Llama 3.2 90B Vision Instruct locally?
Yes — Llama 3.2 90B Vision Instruct can run locally on consumer hardware. At BF16 quantization it needs 194.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Llama 3.2 90B Vision Instruct?
At BF16, Llama 3.2 90B Vision Instruct can reach ~23 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B300 → 8000 ÷ 194.9 × 0.65 = ~27 tok/s
Estimated speed at BF16 (194.9 GB)
~27 tok/s~23 tok/s~23 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Llama 3.2 90B Vision Instruct?
At BF16, the download is about 177.19 GB.
- Which GPUs can run Llama 3.2 90B Vision Instruct?
No single consumer GPU has enough VRAM to run Llama 3.2 90B Vision Instruct at BF16 (194.9 GB). Multi-GPU or professional hardware is required.
- Which devices can run Llama 3.2 90B Vision Instruct?
4 devices with unified memory can run Llama 3.2 90B Vision Instruct at BF16 (194.9 GB), including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA DGX A100 640GB, NVIDIA DGX H100. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.