# How Much RAM Do You Need to Run a Local LLM in 2026? > Source: > Published: 2026-09-02 13:00:00+00:00 **The short answer:** for a small model that fits your GPU, 16GB of system RAM is plenty. For the big Mixture-of-Experts models everyone runs in 2026, the number that matters flipped from VRAM to RAM: you want **at least 64GB, ideally 128GB or more**, because those models keep their rarely-used expert weights in system memory. And RAM *speed* now matters as much as capacity. Here is how to size it for your case. We synthesize this from the file-size math, vendor specs, and owner reports, cited below; we have not benchmarked every configuration first-hand. ## RAM vs VRAM: which one actually gates a local LLM? Two different pools do two different jobs. **VRAM** (on your graphics card) is fast memory the GPU reads directly. **System RAM** is slower but far larger and cheaper per gigabyte. For years the rule was simple: fit the whole model in VRAM or suffer. That rule broke in 2026, because [nearly every notable open model became a sparse Mixture-of-Experts](https://vettedconsumer.com/every-frontier-open-model-is-a-moe-now-what-that-does-to-your-hardware-math/). A tool like llama.cpp can keep an MoE model's hot, always-active layers on the GPU and offload the huge pile of rarely-touched expert weights into system RAM. Suddenly the question is no longer only "how much VRAM," it is "how much RAM." If your budget is a single graphics card rather than a big unified pool, the tiers below still apply: a 16GB card like the [RX 9060 XT 16GB](https://vettedconsumer.com/rx-9060-xt-16gb-buyers-guide-the-budget-value-champ-buy-the-16gb/) sits in the 14B-to-30B-MoE band, and owners who tried stacking cheap cards found that [three RTX 3060s measured against one RTX 3090](https://vettedconsumer.com/three-rtx-3060s-vs-one-rtx-3090-for-local-ai-what-a-1-500-build-actually-measured/) lose on bandwidth, not capacity. On the unified-memory side, [which Mac fits which memory tier](https://vettedconsumer.com/which-mac-for-local-llms-2026-buyers-guide/) is its own decision. ## How much RAM do you need to run a local LLM? Start from the file-size rule (bytes ≈ parameters × bits-per-weight ÷ 8), then add headroom. A rough guide for a 4-bit quant, which is the practical default: | Model you want to run | System RAM to aim for | |---|---| | 8B to 14B (fits most GPUs) | 16GB is fine; the GPU does the work | | 30B-class MoE (gpt-oss-20b, Qwen3-30B-A3B) | 32GB comfortable | | gpt-oss-120b (~63GB at 4-bit) | 64GB minimum, 96GB comfortable | | DeepSeek V4 Flash (284B, ~138GB at 4-bit) | 156GB or more (128GB is not enough) | | GLM-5.2 / Inkling (700B to 1T class) | 256GB+ or unified-memory Mac | That DeepSeek V4 Flash row is not theoretical. A reviewer running the full 284B model on a single RTX 3090 via expert offload found that "128GB is not enough; 156GB probably would be, 168GB more common," exactly the trap our companion coverage of that build documents. The GPU was the easy part; the RAM was the ceiling. ## Does RAM speed matter for local LLMs? Yes, and more than most guides admit. When expert weights stream from system RAM on every token, your [memory bandwidth sets the speed](https://vettedconsumer.com/bandwidth-not-tflops-what-sets-your-local-llm-speed-and-why-the-newest-card-isnt-always-fastest/), just as it does inside a GPU. Slow DDR4-2400 leaves real performance on the table versus DDR4-3200 or DDR5-6000, and dual-channel (or quad-channel) is close to mandatory: a single stick halves your bandwidth and chokes generation. One owner running an offloaded MoE put it simply: "I knew there was a good reason I paid all that money for DDR5 6000." Fill all your memory channels, and buy the faster kit if the budget allows. ## Unified memory changes the math On an Apple Mac or an AMD Strix Halo box, there is no separate VRAM and RAM; it is one [unified pool](https://vettedconsumer.com/unified-memory-explained-why-mini-pcs-can-run-70b-models-a-big-gpu-cant-and-where-they-slow-down/) the GPU reads at high bandwidth. That is why a 128GB Strix Halo mini-PC or a big-memory Mac Studio runs models a 24GB graphics card cannot touch: the whole pool is fast, GPU-accessible memory. If you are buying a machine specifically for local AI, this is the tier to compare, our [128GB matchup](https://vettedconsumer.com/strix-halo-vs-the-mac-for-local-ai-the-128gb-matchup-in-other-peoples-measured-numbers/) covers the tradeoffs. ## The catch: RAM got expensive The uncomfortable part of this advice in 2026 is that [memory prices spiked](https://vettedconsumer.com/why-everything-got-more-expensive-the-memory-crisis-explained-via-dave2d/). The 128GB-plus you now want for MoE offload can cost more than the used GPU you pair it with. Two practical consequences: buy the RAM you need in one go rather than planning to add more later at a worse price, and do the buy-vs-rent math before committing to a giant local build, our [cost calculator](https://vettedconsumer.com/cost-calculator/) prices exactly that. ## The cheat-sheet | Your goal | RAM to buy | |---|---| | Run 8B to 30B models on a GPU | 16 to 32GB, dual-channel | | Offload a 100B-class MoE (GPU + RAM) | 64 to 96GB, fastest kit you can afford | | Run 284B-class models on one GPU + RAM | 156GB+, dual/quad-channel | | Buy one machine for everything | 128GB+ unified memory (Strix Halo or Mac) | The one line to remember: in the MoE era, VRAM decides which models you can run *fast*, but RAM increasingly decides which models you can run *at all*. Size both against your shortlist in our [Can I run it? calculator](https://vettedconsumer.com/can-i-run-it/) before you buy a single stick. ## Sources and how we researched this - The MoE-offload mechanism and active-parameter math: our [MoE-era explainer](https://vettedconsumer.com/every-frontier-open-model-is-a-moe-now-what-that-does-to-your-hardware-math/) , drawing on the[original Mixture-of-Experts paper (Shazeer et al., 2017)](https://arxiv.org/abs/1701.06538?ref=vettedconsumer.com) and llama.cpp's expert-offload documentation. - Model file sizes: the params × bits ÷ 8 rule cross-checked against published GGUF sizes (gpt-oss-120b ~63GB, DeepSeek V4 Flash ~138GB at 4-bit). - Owner RAM findings: attributed reports on running offloaded MoE models, quoted in our related hands-on coverage. We have not tested every configuration first-hand. *Related:* *Every frontier open model is a MoE now* *·* *How much VRAM for a 70B* *·* *Unified memory, explained* *·* *Bandwidth, Not TFLOPS*