Reference
LLM VRAM Requirements
How much memory each model needs to load — checked against the memory your hardware will actually hand it. Those are different numbers, and the gap is why people buy the wrong machine.
Last reviewed 2026-09-22 · Next review 2026-12 · Reviewed quarterly
Memory required, and the smallest box that holds it
| Model | 4-bit (Q4) | 8-bit (Q8) | FP16 | Smallest box that fits (4-bit) |
|---|---|---|---|---|
| Llama 4 Scout 8B8B | 5 GB | 9 GB | 16 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Llama 4 Maverick 70B70B | 40 GB | 75 GB | 140 GB | Apple Mac Mini M5 Pro64GB unified (max M5 Pro config)48GB usableCheck price |
| Llama 4 Behemoth 405B405B | 230 GB | 430 GB | 810 GB | Apple Mac Studio M5 Ultra512GB unified (M5 Ultra max config)384GB usableCheck price |
| DeepSeek R1 7B7B | 4.5 GB | 8 GB | 14 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| DeepSeek R1 70B70B | 40 GB | 75 GB | 140 GB | Apple Mac Mini M5 Pro64GB unified (max M5 Pro config)48GB usableCheck price |
| Gemma 3 9B9B | 6 GB | 10 GB | 18 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Gemma 3 27B27B | 16 GB | 29 GB | 54 GB | Apple Mac Mini M4 Pro24GB unified18GB usableCheck price |
| Qwen 3 7B7B | 4.5 GB | 8 GB | 14 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Qwen 3 72B72B | 42 GB | 77 GB | 144 GB | Apple Mac Mini M5 Pro64GB unified (max M5 Pro config)48GB usableCheck price |
| Stable Diffusion XL3.5B | 4 GB | 6 GB | 8 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Flux.1 Dev12B | 8 GB | 13 GB | 24 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Whisper Large V31.5B | 1.5 GB | 2.5 GB | 4 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Mistral 7B7B | 4.5 GB | 8 GB | 14 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| Phi-4 14B14B | 9 GB | 15 GB | 28 GB | Intel Arc B580 12GB12GB GDDR611GB usableCheck price |
| CodeLlama 34B34B | 20 GB | 36 GB | 68 GB | NVIDIA GeForce RTX 409024GB GDDR6X23GB usableCheck price |
Figures are weights plus loading overhead. "Smallest box that fits" is the lowest-usable-memory item in our catalog that clears the 4-bit figure — not the fastest, and not the cheapest. Check the configuration line — several of these machines ship in memory tiers, and only the configuration named clears the requirement. A buy link lands on the product's listing, which is often the base tier, so you may need to configure the memory up. Datacenter cards (A100, H100) are listed in the table below but excluded from this column: they fit, but they are not a recommendation a normal buyer can act on. Speed is a separate question: see the tokens/sec reference.
Advertised memory vs. memory a model can actually use
The column that matters when you are deciding what to buy is the third one.
| Hardware | Advertised | Usable | Why |
|---|---|---|---|
| Apple Mac Studio M5 Ultra512GB unified (M5 Ultra max config)Check price | 512 GB | 384 GB | macOS reserves a working set from the shared pool. At 512GB even the 25% reservation leaves more usable memory than any other single desktop — this is the only entry here that holds frontier-scale MoE weights. Note Apple ships the 512GB configuration from late October 2026; 256GB is the launch maximum. |
| GMKtec EVO-X2 (Ryzen AI Max+ 395)128GB LPDDR5X (Ryzen AI Max+ 395)Check price | 128 GB | 96 GB | Strix Halo exposes at most ~96GB of its 128GB pool to the GPU — a platform limit, not an OS reservation. This is the gap that catches buyers comparing it to a 128GB spec sheet. |
| NVIDIA DGX Spark128GB LPDDR5X coherent (GB10)Check price | 128 GB | 96 GB | Grace Blackwell coherent memory — CPU and GPU address one pool, so the usable share is high, but the OS still resides in it. |
| Apple Mac Studio M5 Max128GB unified (max M5 Max config)Check price | 128 GB | 96 GB | Same 128GB ceiling as the M4 Max it replaces, so the same models fit. The M5 Max gain is prompt processing, not capacity. |
| Apple Mac Studio M4 Max128GB unified (max M4 Max config)Check price | 128 GB | 96 GB | macOS reserves a working set from the shared pool; GPU-wired memory is capped below total by default. 128GB is the maximum M4 Max configuration (36GB standard, configurable to 48/64/128GB). |
| NVIDIA A100 80GB PCIe80GB HBM2eCheck price | 80 GB | 79 GB | Datacenter card, no display output — reserve is driver context only. |
| NVIDIA H100 PCIe 80GB80GB HBM3Check price | 80 GB | 79 GB | Datacenter card, no display output — reserve is driver context only. |
| Apple Mac Mini M5 Pro64GB unified (max M5 Pro config)Check price | 64 GB | 48 GB | macOS reserves a working set from the shared pool. At 64GB this is the cheapest Apple machine that clears a 70B model at Q4 with context headroom. |
| NVIDIA GeForce RTX 509032GB GDDR7Check price | 32 GB | 31 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 409024GB GDDR6XCheck price | 24 GB | 23 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 309024GB GDDR6XCheck price | 24 GB | 23 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| Apple Mac Mini M4 Pro24GB unifiedCheck price | 24 GB | 18 GB | macOS reserves a working set from the shared pool; GPU-wired memory is capped below total by default. |
| NVIDIA GeForce RTX 508016GB GDDR7Check price | 16 GB | 15 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 5070 Ti16GB GDDR7Check price | 16 GB | 15 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 4080 SUPER16GB GDDR6XCheck price | 16 GB | 15 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 5060 Ti 16GB16GB GDDR7Check price | 16 GB | 15 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| NVIDIA GeForce RTX 4060 Ti 16GB16GB GDDR6Check price | 16 GB | 15 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
| Intel Arc B580 12GB12GB GDDR6Check price | 12 GB | 11 GB | Dedicated VRAM; ~1GB reserved for display + driver context. |
What these numbers exclude
KV cache grows with context. The figures above cover weights plus loading overhead at a short prompt. Every token you keep in context adds key/value cache on top, and at long context that can add many gigabytes. The exact amount depends on the model's layer count, KV-head count and head dimension — architecture details that vary per model. We do not publish a per-model KV table here because we have not verified those configs against primary sources, and a wrong number is worse than no number. Plan headroom above the figure in the table, especially at 32k context and up.
Fitting is not the same as being fast. A model that fits in memory can still run slowly if memory bandwidth is low. Capacity decides whether it runs; bandwidth decides how fast. The tokens/sec reference covers the second question with measured, source-linked figures.
Multi-GPU changes the ceiling, not the math. Two 24GB cards hold a larger model than one, but layer-split inference adds interconnect overhead. Rows marked "needs multi-GPU" are beyond any single item in our catalog.
How the usable figure is calculated
Two rules, applied uniformly, so you can check any row yourself:
- Discrete GPU: usable = VRAM − 1GB (display output and driver/CUDA context).
- Unified memory: usable = total × 0.75 — the OS occupies the same pool. On macOS the GPU-wired limit is raisable, so treat this as a conservative planning figure rather than a hard ceiling.
- Documented hard caps win. Where a platform imposes a lower ceiling than the rule above, that ceiling is used instead — Strix Halo's ~96GB GPU allocation out of a 128GB pool is the case that matters today.
We publish no measured claim on this page. Every figure is either a hardware spec or the output of the arithmetic above. Measured throughput lives on the benchmark reference, where each number cites its source.
Know the model but not the budget? Use the GPU Advisor to match a model and budget to hardware, or browse the full hardware catalog.
Disclosure: product links on this page are affiliate links. We may earn a commission if you make a purchase — at no extra cost to you.