Llama 3.1 405B is the largest open-weight Llama and arguably the highest-quality open base model ever released. At FP16 it requires ~890 GB of VRAM, meaning 8x H100 / B200 minimum; at Q4_K_M it still demands ~245 GB. It's not a homelab model — but quantized aggressively across 4x A100 80 GB it becomes usable for research workloads.
Llama 3.1 405B needs at minimum 245 GB of VRAM at Q4_K_M quantization (891 GB at FP16). The cheapest GPU that comfortably fits with KV-cache headroom is the Apple Mac Studio (M3 Ultra, 512 GB) (512 GB VRAM, $9,499 MSRP). Community benchmark submissions are open. This model exceeds 48 GB at Q4, so plan for a 2-or-more-GPU split.
Weights only. Add ~20-30% for KV cache at typical context lengths.
| Quant | VRAM | Quality | Notes |
|---|---|---|---|
| FP16 | 891.0 GB | Reference | Training-precision weights |
| Q8_0 | 445.5 GB | Near-lossless | 8-bit, ~0.1% perplexity hit |
| Q6_K | 360.9 GB | Very high | 6-bit, near Q8 quality |
| Q5_K_M | 307.4 GB | High | Strong middle ground |
| Q4_K_M | 245.0 GB | Balanced (recommended) | Default for local deployments |
| Q4_0 | 250.6 GB | Legacy 4-bit | Older GGUF, kept for compatibility |
| Q3_K_M | 191.6 GB | Lossy | When VRAM is very tight |
| Q2_K | 142.6 GB | Extreme | Rescue option, quality degrades visibly |
Filtered from MyAIHardware's GPU database. FP16 needs 891.0 GB; Q4_K_M needs 245.0 GB.
| GPU | Per-card VRAM | 2x | 4x | 8x |
|---|---|---|---|---|
| AMD Instinct MI300X | 192 GB | 384 | 768 | 1536 |
| Apple Mac Studio (M2 Ultra, 192 GB) | 192 GB | 384 | 768 | 1536 |
| Apple Mac Pro (M2 Ultra, 192 GB) | 192 GB | 384 | 768 | 1536 |
| NVIDIA B200 | 180 GB | 360 | 720 | 1440 |
| NVIDIA B100 | 180 GB | 360 | 720 | 1440 |
| NVIDIA H200 | 141 GB | 282 | 564 | 1128 |
We don't have community benchmarks for this exact model yet.
Official weights + popular runtime tags.
ollama pull llama3.1:405bStep-by-step guides that use this model.
Models in a similar size or capability class.
Use these in papers, blog posts, or internal docs.
@misc{llama-3-1-405b-2024,
title={Llama 3.1 405B},
author={Meta AI},
year={2024},
url={https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct}
}Meta AI (2024). Llama 3.1 405B [Model card]. Hugging Face. https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct
Llama 3.1 405B (Llama 3.1, Meta AI, 2024). Available at https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct.