L
localai.computer
ModelsGPUsSystemsAI SetupsBuildsOpenClawMethodology

Resources

  • Methodology
  • Submit Benchmark
  • About

Browse

  • AI Models
  • GPUs
  • PC Builds

Guides

  • OpenClaw Guide
  • How-To Guides

Legal

  • Privacy
  • Terms
  • Contact

© 2025 localai.computer. Hardware recommendations for running AI models locally.

ℹ️We earn from qualifying purchases through affiliate links at no extra cost to you. This supports our free content and research.

Can Apple M2 Ultra run RedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamic?

Runs Q4192GB VRAM availableRequires 44GB+

Apple M2 Ultra meets the minimum VRAM requirement for Q4 inference of RedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamic. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.

What this means for you

Apple M2 Ultra can run RedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamic with Q4 quantization. At approximately 22 tokens/second, you can expect Moderate speed - useful for batch processing.

You have 148GB headroom, which is sufficient for system overhead and smooth operation.

Quantization breakdown

QuantizationVRAM neededVRAM availableEstimated speedVerdict
Q444GB192GB21.54 tok/s✅ Fits comfortably
Q888GB192GB14.17 tok/s✅ Fits comfortably
FP16176GB192GB8.79 tok/s✅ Fits comfortably

Best current price

Apple M2 Ultra
$5,999.00 on Amazon
Check Price

Suitable alternatives

AMD Instinct MI300X
192GB
150.03 tok/s
Price: —
NVIDIA H200 SXM 141GB
141GB
137.52 tok/s
Price: —
AMD Instinct MI300X
192GB
96.23 tok/s
Price: —
NVIDIA H100 SXM5 80GB
80GB
90.82 tok/s
Price: —
NVIDIA H200 SXM 141GB
141GB
89.66 tok/s
Price: —

More questions

Apple M2 Ultra specs & pricingFull guide for RedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamicRedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamic speed on Apple M2 UltraRedHatAI/Llama-3.2-90B-Vision-Instruct-FP8-dynamic Q4 requirements