GeForce RTX 4080
NVIDIA
16GB VRAM · 716 GB/s bandwidth · 48.7 FP16 TFLOPS · 320W TDP
The GeForce RTX 4080 has 16GB of VRAM with 716 GB/s memory bandwidth and 48.7 TFLOPS FP16 compute. At Q4 quantization, it can comfortably run Gemma 3 4B (232 tok/s), Qwen 2.5 7B (122 tok/s), Llama 3.1 8B (116 tok/s). Models larger than ~27B parameters won't fit even at Q4. Electricity cost is approximately $35/month at 320W TDP.
What LLMs can you run?
| Model | Params | Q4 Weight | Fit | Decode |
|---|---|---|---|---|
| Gemma 3 4B | 4.0B | 2 GB | comfortable | 232 tok/s |
| Qwen 2.5 7B | 7.6B | 4 GB | comfortable | 122 tok/s |
| Llama 3.1 8B | 8.0B | 4 GB | comfortable | 116 tok/s |
| Mistral Small 24B | 24.0B | 12 GB | comfortable | 38 tok/s |
| Gemma 3 27B | 27.4B | 14 GB | won't fit | |
| Qwen 2.5 Coder 32B | 32.5B | 16 GB | won't fit | |
| Llama 3.3 70B | 70.6B | 35 GB | won't fit | |
| Qwen 2.5 72B | 72.7B | 36 GB | won't fit | |
| Llama 3.1 405B | 405B | 202 GB | won't fit | |
| DeepSeek R1 671B | 671B | 336 GB | won't fit |
Similar GPUs
| GPU | VRAM | BW | TFLOPS | TDP |
|---|---|---|---|---|
| Quadro GP100 | 16GB | 732 GB/s | 20.7 | 235W |
| Tesla P100 DGXS | 16GB | 732 GB/s | 21.2 | 300W |
| Tesla P100 PCIe 16 GB | 16GB | 732 GB/s | 19.1 | 250W |
| Tesla P100 SXM2 | 16GB | 732 GB/s | 21.2 | 300W |
| GeForce RTX 4080 SUPER | 16GB | 736 GB/s | 52.2 | 320W |
Compare with another GPU
Select another GPU to compare specs and model performance side by side.