⚑ LLMVRAMCalc.com
EN δΈ­ζ–‡ ES DE
Calculadora VRAM β€Ί GPUs β€Ί 2x NVIDIA RTX 4090 (48GB)

What LLMs Run on 2x NVIDIA RTX 4090 (48GB) (48GB VRAM)?

2x NVIDIA RTX 4090 (48GB) (48GB VRAM, 1008 GB/s bandwidth, 330.4 TFLOPS FP16).

Modelo ParΓ‘metros (Activos) Q4_K_M (4-bit) VRAM FP16 Status on 2x NVIDIA RTX 4090 (48GB) Decode / Prefill Report
DeepSeek-R1 671B (MoE) 671B 378.7 GB 1250.87 GB No ~3 / 2322 tok/s Calculate β†’
DeepSeek-V3-0324 671B (MoE) 671B 378.7 GB 1250.87 GB No ~3 / 2322 tok/s Calculate β†’
DeepSeek-R1-Distill-Llama-70B 70.55B 43.09 GB 134.41 GB Fits Q4_K_M ~16.3 / 1218 tok/s Calculate β†’
DeepSeek-R1-Distill-Qwen-32B 32.76B 21.23 GB 63.51 GB Fits Q4_K_M ~35 / 2622 tok/s Calculate β†’
DeepSeek-R1-Distill-Qwen-14B 14.77B 10.61 GB 29.5 GB FP16 & Q4 ~77.7 / 5816 tok/s Calculate β†’
DeepSeek-R1-Distill-Llama-8B 8.03B 6.22 GB 16.45 GB FP16 & Q4 ~142.9 / 10698 tok/s Calculate β†’
DeepSeek-R1-Distill-Qwen-7B 7.62B 5.43 GB 15.12 GB FP16 & Q4 ~150.6 / 11273 tok/s Calculate β†’
Qwen3-235B-A22B (MoE) 235.1B 134.4 GB 439.87 GB No ~6.5 / 3870 tok/s Calculate β†’
Qwen3-30B-A3B (MoE) 30.53B 18.53 GB 58.1 GB Fits Q4_K_M ~343.5 / 18000 tok/s Calculate β†’
Qwen3-32B 32.76B 21.22 GB 63.51 GB Fits Q4_K_M ~35 / 2622 tok/s Calculate β†’
Qwen3-14B 14.77B 10.36 GB 29.25 GB FP16 & Q4 ~77.7 / 5816 tok/s Calculate β†’
Qwen3-8B 8.19B 6.47 GB 16.87 GB FP16 & Q4 ~140.1 / 10489 tok/s Calculate β†’
Qwen3-4B 4.02B 3.61 GB 8.68 GB FP16 & Q4 ~285.4 / 18000 tok/s Calculate β†’
Qwen2.5-72B-Instruct 72.71B 44.38 GB 138.43 GB Fits Q4_K_M ~15.8 / 1181 tok/s Calculate β†’
Qwen2.5-Coder-32B-Instruct 32.76B 21.23 GB 63.51 GB Fits Q4_K_M ~35 / 2622 tok/s Calculate β†’
Qwen2.5-14B-Instruct 14.77B 10.61 GB 29.5 GB FP16 & Q4 ~77.7 / 5816 tok/s Calculate β†’
Qwen2.5-7B-Instruct 7.62B 5.43 GB 15.12 GB FP16 & Q4 ~150.6 / 11273 tok/s Calculate β†’
QwQ-32B Reasoning 32.76B 21.23 GB 63.51 GB Fits Q4_K_M ~35 / 2622 tok/s Calculate β†’
Llama-4-Maverick-17B-128E (400B MoE) 400B 227.32 GB 747.05 GB No ~7 / 4994 tok/s Calculate β†’
Llama-4-Scout-17B-16E (109B MoE) 109B 63.7 GB 205.02 GB No ~18.3 / 4994 tok/s Calculate β†’
Llama-3.3-70B-Instruct 70.55B 43.09 GB 134.41 GB Fits Q4_K_M ~16.3 / 1218 tok/s Calculate β†’
Llama-3.1-405B-Instruct 405B 233.02 GB 758.82 GB No ~1.2 / 212 tok/s Calculate β†’
Llama-3.1-70B-Instruct 70.55B 43.09 GB 134.41 GB Fits Q4_K_M ~16.3 / 1218 tok/s Calculate β†’
Llama-3.1-8B-Instruct 8.03B 6.22 GB 16.45 GB FP16 & Q4 ~142.9 / 10698 tok/s Calculate β†’
Llama-3.2-3B-Instruct 3.21B 3.33 GB 7.34 GB FP16 & Q4 ~350 / 18000 tok/s Calculate β†’
Mistral-Large-2-123B 122.6B 72.84 GB 231.62 GB No ~2 / 701 tok/s Calculate β†’
Mistral-Small-3.1-24B-Instruct 23.6B 15.28 GB 45.7 GB FP16 & Q4 ~48.6 / 3640 tok/s Calculate β†’
Mixtral-8x22B-Instruct (MoE) 140.6B 81.38 GB 264.13 GB No ~5.4 / 2197 tok/s Calculate β†’
Mixtral-8x7B-Instruct (MoE) 46.7B 27.8 GB 88.47 GB Fits Q4_K_M ~88.9 / 6659 tok/s Calculate β†’
Codestral-22B 22.2B 14.81 GB 43.59 GB FP16 & Q4 ~51.7 / 3870 tok/s Calculate β†’
Google Gemma-3-27B-IT 27.4B 20.34 GB 55.4 GB Fits Q4_K_M ~41.9 / 3135 tok/s Calculate β†’
Google Gemma-3-12B-IT 12.2B 10.75 GB 26.21 GB FP16 & Q4 ~94 / 7041 tok/s Calculate β†’
Google Gemma-3-4B-IT 4.3B 4.24 GB 9.56 GB FP16 & Q4 ~266.8 / 18000 tok/s Calculate β†’
Microsoft Phi-4 14B 14.66B 10.5 GB 29.36 GB FP16 & Q4 ~78.3 / 5860 tok/s Calculate β†’
Cohere Command-A 111B 111B 66.19 GB 209.26 GB No ~2.6 / 774 tok/s Calculate β†’
GLM-4-32B-0414 32B 20.77 GB 62 GB Fits Q4_K_M ~35.9 / 2685 tok/s Calculate β†’