AMD Radeon AI PRO R9700 vs NVIDIA RTX PRO 6000 Blackwell
For running models locally, priced in Canada. At 128k context the first runs 24 models fully in memory, the second 39.
In short
The NVIDIA RTX PRO 6000 Blackwell has more memory for models: 96 GB against 32 GB. At 128k context the NVIDIA RTX PRO 6000 Blackwell runs 39 of the models we track fully in memory and the AMD Radeon AI PRO R9700 24; only the NVIDIA RTX PRO 6000 Blackwell holds Qwen 3.8 Flash Next. On the models both run fully in memory, the NVIDIA RTX PRO 6000 Blackwell is about 3.0× as fast as the AMD Radeon AI PRO R9700 (estimated, median across models). In Canada the AMD Radeon AI PRO R9700 costs $2,800 at Amazon and the NVIDIA RTX PRO 6000 Blackwell $3,133 at Amazon: the AMD Radeon AI PRO R9700 is $333 cheaper.
Updated . Speeds are estimates; how they are worked out.
Specifications
| Spec | AMD Radeon AI PRO R9700 | NVIDIA RTX PRO 6000 Blackwell |
|---|---|---|
| GPU | Radeon AI PRO R9700 | RTX PRO 6000 Blackwell |
| Memory | 32 GB GDDR6 | 96 GB GDDR7 |
| Memory bus | 256-bit | 512-bit |
| Memory bandwidth | 640 GB/s | 1,792 GB/s |
| FP16 compute | 191 TFLOPS | 504 TFLOPS |
| Software | ROCm | CUDA |
| Board power | 300 W | 600 W |
| PCIe generation | Gen 5 | Gen 5 |
| PCIe lanes | x16 | x16 |
What each runs
Estimated decode speed with the most precise quant that fits each one est.
| Model | AMD Radeon AI PRO R9700 | NVIDIA RTX PRO 6000 Blackwell |
|---|---|---|
| Qwen 3.8 Flash Next | won't run | 20tok/sUD-Q3_K_XL |
| Qwen 3.8 27B | 12tok/sUD-Q6_K | 29tok/sQ8_0 |
| Qwen 3.5 122B-A10B | offload | 62tok/sQ5_K_M |
| Qwen 3.6 27B | 11tok/sQ6_K | 30tok/sQ8_0 |
| Gemma 4 31B | 12tok/sQ4_K_M | 25tok/sQ8_0 |
| Qwen3 Coder Next | offload | 90tok/sQ8_0 |
| Nemotron 3 Super 120B-A12B | offload | 72tok/sUD-Q4_K_M |
| gpt-oss 120B | won't run | 84tok/sMXFP4 |
| Mistral Small 4 119B | offload | 80tok/sUD-Q5_K_M |
| Qwen 3.6 35B-A3B | 43tok/sUD-Q5_K_M | 114tok/sQ8_0 |
| Qwen 3.5 27B | 11tok/sQ6_K | 30tok/sQ8_0 |
| Gemma 4 26B-A4B | 30tok/sQ8_0 | 95tok/sQ8_0 |
| GLM 4.5 Air | won't run | 19tok/sQ4_K_M |
| Qwen 3.5 35B-A3B | 41tok/sQ6_K | 114tok/sQ8_0 |
| GLM 4.7 Flash | 19tok/sQ6_K | 54tok/sQ8_0 |
| Qwen3 Next 80B-A3B | offload | 104tok/sQ8_0 |
| Devstral Small 2 24B | offload | 24tok/sQ8_0 |
| Nemotron 3.5 Lightning 30B-A3B | offload | 160tok/sUD-Q6_K_XL |
| Kimi Linear 48B-A3B | 72tok/sQ4_K_M | 155tok/sQ8_0 |
| gpt-oss 20B | 37tok/sMXFP4 | 118tok/sMXFP4 |
| Qwen3 Coder 30B-A3B | 14tok/sQ4_K_M | 42tok/sQ8_0 |
| Qwen3 32B | offload | 16tok/sQ8_0 |
| Gemma 4 12B | 23tok/sQ8_0 | 73tok/sQ8_0 |
| Qwen 3.5 9B | 25tok/sQ8_0 | 80tok/sQ8_0 |
| Gemma 3 27B | 12tok/sQ5_K_M | 28tok/sQ8_0 |
| Mistral Small 3.2 24B | offload | 24tok/sQ8_0 |
| DeepSeek R1 Distill Qwen 32B | offload | 16tok/sQ8_0 |
| Llama 3.3 70B | won't run | 12tok/sQ5_K_M |
| Qwen3 14B | 12tok/sQ4_K_M | 30tok/sQ8_0 |
| Ministral 3 14B | 11tok/sQ5_K_M | 31tok/sQ8_0 |
| DeepSeek R1 0528 Qwen3 8B | 13tok/sQ8_0 | 40tok/sQ8_0 |
| Qwen3 8B | 13tok/sQ8_0 | 40tok/sQ8_0 |
| Phi-4 14B | offload | 26tok/sQ8_0 |
| Qwen 3.5 4B | 40tok/sQ8_0 | 126tok/sQ8_0 |
| Ministral 3 8B | 13tok/sQ8_0 | 41tok/sQ8_0 |
| Granite 4.2 8B | 11tok/sQ8_0 | 36tok/sQ8_0 |
| Gemma 4 E4B | 29tok/sQ8_0 | 93tok/sQ8_0 |
| LFM2.5 8B-A1B | 28tok/sQ8_0 | 89tok/sQ8_0 |
| Llama 3.1 8B | 15tok/sQ6_K | 47tok/sQ6_K |
| DeepSeek V4 Flash | won't run | offload |
| GLM 5.3 Flash | won't run | offload |
| MiniMax M2.7 | won't run | offload |
Questions
- Which is better for local AI, the AMD Radeon AI PRO R9700 or the NVIDIA RTX PRO 6000 Blackwell?
- Choose the NVIDIA RTX PRO 6000 Blackwell to run bigger models. Choose the NVIDIA RTX PRO 6000 Blackwell for speed. Choose the AMD Radeon AI PRO R9700 to spend less.
- Which can run bigger models, the AMD Radeon AI PRO R9700 or the NVIDIA RTX PRO 6000 Blackwell?
- At 128k context the NVIDIA RTX PRO 6000 Blackwell runs 39 of the models we track fully in memory and the AMD Radeon AI PRO R9700 24; only the NVIDIA RTX PRO 6000 Blackwell holds Qwen 3.8 Flash Next.
- Which is faster for local AI, the AMD Radeon AI PRO R9700 or the NVIDIA RTX PRO 6000 Blackwell?
- On the models both run fully in memory, the NVIDIA RTX PRO 6000 Blackwell is about 3.0× as fast as the AMD Radeon AI PRO R9700 (estimated, median across models).
- What is the price difference between the AMD Radeon AI PRO R9700 and the NVIDIA RTX PRO 6000 Blackwell?
- In Canada the AMD Radeon AI PRO R9700 costs $2,800 at Amazon and the NVIDIA RTX PRO 6000 Blackwell $3,133 at Amazon: the AMD Radeon AI PRO R9700 is $333 cheaper.
More comparisons
- Intel Arc Pro B60 vs AMD Radeon AI PRO R9700
- Mac mini M4 32GB vs AMD Radeon AI PRO R9700
- Mac Studio M3 Ultra 96GB vs NVIDIA RTX PRO 6000 Blackwell
- AMD Radeon AI PRO R9700 vs NVIDIA GeForce RTX 3090
- AMD Radeon AI PRO R9700 vs NVIDIA GeForce RTX 3090 Ti
- AMD Radeon AI PRO R9700 vs NVIDIA GeForce RTX 5080
- AMD Radeon AI PRO R9700 vs NVIDIA GeForce RTX 5090
- AMD Radeon AI PRO R9700 vs AMD Radeon RX 7900 XTX
- NVIDIA GeForce RTX 3090 vs NVIDIA RTX PRO 6000 Blackwell
- NVIDIA GeForce RTX 3090 Ti vs NVIDIA RTX PRO 6000 Blackwell