Quadro RTX 8000 vs RTX A6000 for local AI
Both fit the same models we track, up to Llama 3.3 70B. The RTX A6000 generates about 14% faster on any model both can run, because it has 768 GB/s of memory bandwidth to the Quadro RTX 8000's 672 GB/s.
Memory and bandwidth
Memory decides which models fit. Bandwidth decides how fast they generate, because every token reads the model's active weights once.
| Quadro RTX 8000 | RTX A6000 | |
|---|---|---|
| Memory | 48 GB VRAM | 48 GB VRAM |
| Bandwidth | 672 GB/s | 768 GB/s |
| Biggest model that fits | Llama 3.3 70B | Llama 3.3 70B |
| Best everyday pick | Qwen3 30B-A3B, ~120 tok/s | Qwen3 32B, ~21 tok/s |
Every model, side by side
Estimated generation speed at the largest memory size each was sold in (48 GB and 48 GB).
| Model | Quadro RTX 8000 | RTX A6000 |
|---|---|---|
| Qwen3 4B About 4 GB | ~150 tok/s Runs great | ~170 tok/s Runs great |
| Qwen3 8B About 7 GB | ~71 tok/s Runs great | ~81 tok/s Runs great |
| Gemma 3 12B About 10 GB | ~46 tok/s Runs great | ~52 tok/s Runs great |
| Qwen3 14B About 11 GB | ~40 tok/s Runs great | ~45 tok/s Runs great |
| gpt-oss 20B About 16 GB | ~120 tok/s Runs great | ~130 tok/s Runs great |
| Gemma 3 27B About 19 GB | ~22 tok/s Usable | ~25 tok/s Usable |
| Qwen3 30B-A3B About 21 GB | ~120 tok/s Runs great | ~130 tok/s Runs great |
| Qwen3 32B About 22 GB | ~18 tok/s Usable | ~21 tok/s Usable |
| Llama 3.3 70B About 45 GB | ~9 tok/s Slow | ~10 tok/s Slow |
| gpt-oss 120B About 67 GB | Needs 67 GB | Needs 67 GB |
| Qwen3 235B-A22B About 144 GB | Needs 144 GB | Needs 144 GB |
| DeepSeek R1 671B About 406 GB | Needs 406 GB | Needs 406 GB |
The RTX A6000, for speed. They fit the same models. Price matters as much as either number: check used prices on the builds page, where every parts list has a measured or estimated speed.
Both columns use the same formula: memory bandwidth divided by the weights read per token, calibrated on our own rack, with Ollama's default 4-bit sizes. Real results are usually within about a third. Measured either one? Send us the number.