Source: https://getrigbase.com/can-it-run/ Last updated: 2026-11-03 # Can I run it? VRAM & KV cache calculator Quick answer **How much VRAM do I need to run an LLM locally?** Weights + KV cache + runtime overhead must fit in the GPU memory. Weights ≈ parameters × bits per weight ÷ 8 (Q4_K_M is 4.89 bits per weight). The KV cache grows with context: 2 × layers × KV heads × head_dim × context tokens × bytes per value; for Llama 3.3 70B at FP16 that is about 0.33 MB per token, 10.7 GB at 32,768 tokens. The calculator below does this math for 1–8 GPUs, DGX Spark, Mac and Strix Halo, and links the measured builds for each setup. Last updated: 3 Nov 2026 Will a 2026 model fit on your 1–8 GPUs, a DGX Spark (or two), a Mac or a Strix Halo? Pick the model, engine and format (GGUF, AWQ, GPTQ, FP8, NVFP4, MXFP4, EXL, MLX), context and hardware. We show the weights, the KV cache, the math, and the real measured builds for that setup. The model decides how many parameters must sit in memory. For **MoE** models (e.g. “35B-A3B”) all experts must be loaded, even though only the active ones are used per token. Total parameters in billions. Weights (GB) = params × bits-per-weight ÷ 8. Edit it for a model we do not list.