A800 PCIe 40 GB
NVIDIA
40GB VRAM · 1560 GB/s bandwidth · 78.0 FP16 TFLOPS · 250W TDP
The A800 PCIe 40 GB has 40GB of VRAM with 1560 GB/s memory bandwidth and 78.0 TFLOPS FP16 compute. At Q4 quantization, it can comfortably run Gemma 3 4B (483 tok/s), Qwen 2.5 7B (254 tok/s), Llama 3.1 8B (240 tok/s). Models larger than ~68B parameters won't fit even at Q4. Electricity cost is approximately $27/month at 250W TDP.
What LLMs can you run?
| Model | Params | Q4 Weight | Fit | Decode |
|---|---|---|---|---|
| Gemma 3 4B | 4.0B | 2 GB | comfortable | 483 tok/s |
| Qwen 2.5 7B | 7.6B | 4 GB | comfortable | 254 tok/s |
| Llama 3.1 8B | 8.0B | 4 GB | comfortable | 240 tok/s |
| Mistral Small 24B | 24.0B | 12 GB | comfortable | 80 tok/s |
| Gemma 3 27B | 27.4B | 14 GB | comfortable | 70 tok/s |
| Qwen 2.5 Coder 32B | 32.5B | 16 GB | comfortable | 59 tok/s |
| 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 |
|---|---|---|---|---|
| A100 PCIe 40 GB | 40GB | 1560 GB/s | 78.0 | 250W |
| A100 SXM4 40 GB | 40GB | 1560 GB/s | 78.0 | 400W |
| M4 Max 36GB | 36GB | 273 GB/s | 36.9 | 75W |
| M3 Max 36GB | 36GB | 200 GB/s | 32.8 | 75W |
| M3 Pro 36GB | 36GB | 150 GB/s | 14.8 | 30W |
Compare with another GPU
Select another GPU to compare specs and model performance side by side.