Intel Arc Pro B70 vs AMD Radeon AI PRO R9700
For running models locally, priced in Germany. At 32k context the first runs 61 models fully in memory, the second 61.
In short
Both have 32 GB of memory for models; the Intel Arc Pro B70 has 608 GB/s of bandwidth and the AMD Radeon AI PRO R9700 640 GB/s. At 32k context both run the same 61 models we track fully in memory. On the models both run fully in memory, the AMD Radeon AI PRO R9700 is about 1.2× as fast as the Intel Arc Pro B70 (estimated, median across models).
Updated . Speeds are estimates; how they are worked out.
Specifications
| Spec | Intel Arc Pro B70 | AMD Radeon AI PRO R9700 |
|---|---|---|
| GPU | Arc Pro B70 | Radeon AI PRO R9700 |
| Memory | 32 GB GDDR6 | 32 GB GDDR6 |
| Memory bus | 256-bit | 256-bit |
| Memory bandwidth | 608 GB/s | 640 GB/s |
| FP16 compute | 45.8 TFLOPS | 191 TFLOPS |
| Software | SYCL / oneAPI | ROCm |
| Board power | 230 W | 300 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 | Intel Arc Pro B70 | AMD Radeon AI PRO R9700 |
|---|---|---|
| Qwen 3.8 27B | 10tok/sQ8_0 | 11tok/sQ8_0 |
| Qwen 3.6 27B | 10tok/sQ8_0 | 11tok/sQ8_0 |
| Gemma 4 31B | 11tok/sQ6_K | 12tok/sQ6_K |
| Qwen 3.6 35B-A3B | 50tok/sQ6_K | 64tok/sQ6_K |
| Qwen 3.5 27B | 10tok/sQ8_0 | 11tok/sQ8_0 |
| Gemma 4 26B-A4B | 33tok/sQ8_0 | 42tok/sQ8_0 |
| Qwen 3.5 35B-A3B | 53tok/sQ6_K | 67tok/sQ6_K |
| GLM 4.7 Flash | 29tok/sQ6_K | 36tok/sQ6_K |
| Devstral Small 2 24B | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Nemotron 3.5 Lightning 30B-A3B | 54tok/sUD-Q5_K_M | 68tok/sUD-Q5_K_M |
| Kimi Linear 48B-A3B | 78tok/sQ4_K_M | 99tok/sQ4_K_M |
| gpt-oss 20B | 52tok/sMXFP4 | 65tok/sMXFP4 |
| Qwen3 Coder 30B-A3B | 29tok/sQ6_K | 37tok/sQ6_K |
| Qwen3 32B | 10tok/sQ5_K_M | 11tok/sQ5_K_M |
| Gemma 4 12B | 22tok/sQ8_0 | 26tok/sQ8_0 |
| Qwen 3.5 9B | 29tok/sQ8_0 | 33tok/sQ8_0 |
| Mistral Small 3.2 24B | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Gemma 3 27B | 10tok/sQ8_0 | 11tok/sQ8_0 |
| DeepSeek R1 Distill Qwen 32B | 10tok/sQ5_K_M | 11tok/sQ5_K_M |
| Ministral 3 14B | 15tok/sQ8_0 | 18tok/sQ8_0 |
| Qwen3 14B | 14tok/sQ8_0 | 17tok/sQ8_0 |
| DeepSeek R1 0528 Qwen3 8B | 22tok/sQ8_0 | 26tok/sQ8_0 |
| Qwen3 8B | 22tok/sQ8_0 | 26tok/sQ8_0 |
| Phi-4 14B | 14tok/sQ8_0 | 16tok/sQ8_0 |
| Qwen 3.5 4B | 54tok/sQ8_0 | 63tok/sQ8_0 |
| Ministral 3 8B | 22tok/sQ8_0 | 26tok/sQ8_0 |
| Granite 4.2 8B | 21tok/sQ8_0 | 24tok/sQ8_0 |
| Gemma 4 E4B | 33tok/sQ8_0 | 38tok/sQ8_0 |
| LFM2.5 8B-A1B | 63tok/sQ8_0 | 79tok/sQ8_0 |
| Llama 3.1 8B | 28tok/sQ6_K | 32tok/sQ6_K |
| Qwen2.5 0.5B | 326tok/sQ8_0 | 377tok/sQ8_0 |
| Qwen3.5 0.8B | 247tok/sQ8_0 | 285tok/sQ8_0 |
| Qwen3.5 2B | 164tok/sQ5_K_M | 190tok/sQ5_K_M |
| DeepSeek R1 Distill Qwen 1.5B | 107tok/sQ8_0 | 124tok/sQ8_0 |
| MiniCPM5 2B | 74tok/sQ8_0 | 86tok/sQ8_0 |
| Spark X2.5 4B | 71tok/smlx-4bit | 82tok/smlx-4bit |
| LFM2.5 2.6B | 62tok/sQ8_0 | 72tok/sQ8_0 |
| LFM2.5 VL 3B | 62tok/sQ8_0 | 72tok/sQ8_0 |
| FrogNano 4B 2609 | 53tok/sQ8_0 | 61tok/sQ8_0 |
| Mellum2.1 12B A2.5B Thinking | 53tok/sQ8_0 | 67tok/sQ8_0 |
| granite vision 4.1 4b | 48tok/sQ8_0 | 56tok/sQ8_0 |
| granite 4.2 3b | 46tok/sQ8_0 | 54tok/sQ8_0 |
| Ministral 3 3B Instruct 2512 | 43tok/sQ8_0 | 49tok/sQ8_0 |
| Kimi VL A3B Thinking 2506 | 41tok/sQ8_0 | 51tok/sQ8_0 |
| Kimi VL A3B | 39tok/smlx-8bit | 49tok/smlx-8bit |
| LFM2.5 24B A2B | 35tok/sQ8_0 | 44tok/sQ8_0 |
| GLM 4 9B 0414 | 32tok/smlx-5bit | 37tok/smlx-5bit |
| DeepSeek R1 Distill Qwen 7B | 31tok/sQ8_0 | 35tok/sQ8_0 |
| GLM 4.6V Flash | 27tok/sQ8_0 | 31tok/sQ8_0 |
| DeepSeek R1 Distill Llama 8B | 24tok/sQ8_0 | 27tok/sQ8_0 |
| GLM 4.1V 9B Thinking | 23tok/smlx-8bit | 27tok/smlx-8bit |
| Ministral 3 8B Reasoning 2512 | 22tok/sQ8_0 | 26tok/sQ8_0 |
| Ministral 3 14B Reasoning 2512 | 15tok/sQ8_0 | 18tok/sQ8_0 |
| DeepSeek R1 Distill Qwen 14B | 14tok/sQ8_0 | 16tok/sQ8_0 |
| Qwen3.8 27B OBLITERATED | 13tok/smlx-4bit | 15tok/smlx-4bit |
| Muse Glimmer 30B | 13tok/sQ6_K | 15tok/sQ6_K |
| granite 4.2 30b | 10tok/sQ5_K_M | 12tok/sQ5_K_M |
| Devstral Small 2507 | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Magistral Small 2506 | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Magistral Small 2507 | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Magistral Small 2509 | 10tok/sQ8_0 | 12tok/sQ8_0 |
| Qwen 3.5 122B-A10B | offload | offload |
| Qwen3 Coder Next | offload | offload |
| Nemotron 3 Super 120B-A12B | offload | offload |
| Mistral Small 4 119B | offload | offload |
| GLM 4.5 Air | offload | offload |
| Qwen3 Next 80B-A3B | offload | offload |
| Llama 3.3 70B | offload | offload |
| DeepSeek R1 Distill Llama 70B | offload | offload |
| DeepSeek V4 Flash | won't run | won't run |
| GLM 5.3 Flash | won't run | won't run |
| Qwen 3.8 Flash Next | won't run | won't run |
| MiniMax M2.7 | won't run | won't run |
| gpt-oss 120B | won't run | won't run |
| DeepSeek R1 | won't run | won't run |
| DeepSeek R1 0528 | won't run | won't run |
| DeepSeek R1 Zero | won't run | won't run |
| DeepSeek V3 | won't run | won't run |
| DeepSeek V3 0324 | won't run | won't run |
| DeepSeek V3.1 | won't run | won't run |
| DeepSeek V3.1 Terminus | won't run | won't run |
| DeepSeek V3.2 | won't run | won't run |
| DeepSeek V4 Flash 0731 | won't run | won't run |
| DeepSeek V4 Flash Vision Exp | won't run | won't run |
| Devstral 2 123B Instruct 2512 | won't run | won't run |
| GLM 4.5 | won't run | won't run |
| GLM 4.5V | won't run | won't run |
| GLM 4.6 | won't run | won't run |
| GLM 4.6V | won't run | won't run |
| GLM 4.7 | won't run | won't run |
| GLM 5 | won't run | won't run |
| GLM 5.1 | won't run | won't run |
| GLM 5.2 | won't run | won't run |
| GLM 5.3 | won't run | won't run |
| Kimi Dev 72B | won't run | won't run |
| Kimi K2 | won't run | won't run |
| Kimi K2 Instruct 0905 | won't run | won't run |
| Kimi K2 Thinking | won't run | won't run |
| Kimi K2.5 | won't run | won't run |
| MiniMax M2 | won't run | won't run |
| MiniMax M2.1 | won't run | won't run |
| MiniMax M2.5 | won't run | won't run |
| MiniMax M3 | won't run | won't run |
| Mistral Medium 3.5 128B | won't run | won't run |
Questions
- Which is better for local AI, the Intel Arc Pro B70 or the AMD Radeon AI PRO R9700?
- Choose the AMD Radeon AI PRO R9700 for speed.
- Which can run bigger models, the Intel Arc Pro B70 or the AMD Radeon AI PRO R9700?
- At 32k context both run the same 61 models we track fully in memory.
- Which is faster for local AI, the Intel Arc Pro B70 or the AMD Radeon AI PRO R9700?
- On the models both run fully in memory, the AMD Radeon AI PRO R9700 is about 1.2× as fast as the Intel Arc Pro B70 (estimated, median across models).
More comparisons
- Intel Arc Pro B60 vs Intel Arc Pro B70
- Intel Arc Pro B70 vs NVIDIA RTX PRO 4000 Blackwell
- Mac mini M4 32GB vs AMD Radeon AI PRO R9700
- Mac mini M4 Pro 48GB vs AMD Radeon AI PRO R9700
- AMD Radeon AI PRO R9700 vs AMD Radeon PRO W7900
- 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 4080 Super
- AMD Radeon AI PRO R9700 vs NVIDIA GeForce RTX 5090
- AMD Radeon AI PRO R9700 vs NVIDIA RTX PRO 4000 Blackwell