What GPU do you need to run Mistral Large 4?
No single GPU. Mistral Large 4 needs about 606.5 GB at UD-Q4_K_XL, more than any GPU, system or build we price.
1050B parameters, 52B active (mixture-of-experts) (Mistral docs, accessed 2026-10-29). UD-Q4_K_XL GGUF: 605 GB (source). No context estimate: Mistral Large 4's DSA sparse attention, with KV shared across layers, has no verified llama.cpp KV formula yet (details).
Mistral's docs give 1.05T total and 52B active parameters. The Hugging Face model card says about 1T total and 49B active per token, or 52B counting the embedding and output layers, so both active figures are right. No plain Q4_K_M size from Unsloth, so the 4-bit size is Unsloth's UD-Q4_K_XL (604.8 GB); its smallest file is UD-IQ1_S at 233 GB. Measured runs are slower than our formula: Hugging Face measured about 15 tok/s on a 512 GB M3 Ultra (Q2_K) and 30 tok/s on four RTX Pro 6000s (IQ2_XXS, short prompts), where our formula gives 32 to 39 and about 89, so read any speed estimate for this model as roughly 2 to 3 times too high. Running it yourself is allowed for commercial use; a business offering the model as a service to third parties needs a separate license from Mistral.
Other places to get it
Checked Nov 14, 2026. We list only copies we opened and checked. Check file hashes for copies from other uploaders.
- ModelScope: Other uploaderAccount named Mistral AI; we found no Mistral page linking to it
- Ollama: Cloud only (runs on Ollama's servers)
ollama run mistral-large-4:cloud
Weights-only fits, context unknown. We have no KV cache data for Mistral Large 4, so every fit on this page only checks that the UD-Q4_K_XL file and headroom fit. How much context fits beside the weights is unknown, so none of these is a proven fit for any context length.
No single GPU in our database fits it. Use multiple GPUs (see calculator).
Speed calculator
estimateEstimate = bandwidth ÷ GB read per token (the model's listed GGUF file: Q4_K_M for most models, otherwise the quant named in the list; scaled by active/total parameters for MoE), scaled by a per-platform efficiency and fixed per-token costs calibrated to 45 measured llama.cpp benchmarks (2025–26). Fit = file size + 1.5 GB headroom ≤ total VRAM (on Macs, the part of unified memory macOS lets the GPU use by default), and room for at least 4K tokens of context with an 8-bit KV cache. For a model with no KV cache data, only the file and headroom are checked, so a fit says nothing about context. Multi-GPU: VRAM adds, speed stays one card's (a layer split runs the cards in turn); max context uses the total. Speed is for short context. Max context = what is left after the file and a 1 GB runtime reserve ÷ KV cache bytes per token, default FP16 and 8-bit (q8_0) cache. How speed works · how context works