Intel graphics card
What AI models can the Intel Arc Pro B70 32 GB run?
At 4-bit quantization with an 8K-token conversation, 61 of the 146 open-weight models in our directory fit entirely in the Intel Arc Pro B70 32 GB's 32 GB of video memory (about 31 GB usable), up to Skyfall 36B V2. With 32 GB of dual-channel DDR5-5600, 11 more run partly from system RAM, more slowly. Among models of 7B parameters or more, Laguna XS 2.1 (free) is the fastest, at an estimated 66 to 86 tokens per second.
| Model | Parameters | Memory needed | Estimated speed |
|---|---|---|---|
| Runs entirely in fast memory (61) | |||
| Skyfall 36B V2 Thedrummer | 36.9B | 25 GB 22 GB + 2.1 GB cache | 14–18 tokens/s Comfortable to read |
| Qwen3.5 35B-A3B Alibaba (Qwen) | 36.0B 4.7B active | 23 GB 22 GB + 0.2 GB cache | 40–51 tokens/s Fast |
| Qwen3.6 35B-A3B Alibaba (Qwen) | 36.0B 4.7B active | 23 GB 22 GB + 0.2 GB cache | 40–51 tokens/s Fast |
| Nex-N2.5-Mini NEX AGI | 35.1B 3.9B active | 22 GB 21 GB + 0.2 GB cache | 47–60 tokens/s Fast |
| Laguna XS 2.1 (free) Poolside | 33.4B 2.2B active | 21 GB 20 GB + 0.4 GB cache | 66–86 tokens/s Fast |
| Qwen3 VL 32B Instruct Alibaba (Qwen) | 33.4B | 23 GB 20 GB + 2.1 GB cache | 15–19 tokens/s Comfortable to read |
| Qwen2.5 Coder 32B Instruct Alibaba (Qwen) | 32.8B | 23 GB 20 GB + 2.1 GB cache | 15–20 tokens/s Comfortable to read |
| Qwen3 32B Alibaba (Qwen) | 32.8B | 23 GB 20 GB + 2.1 GB cache | 15–20 tokens/s Comfortable to read |
| Nemotron 3 Nano 30B A3B NVIDIA | 31.6B | 20 GB 19 GB + 0.1 GB cache | Speed not estimated for this architecture |
| Nemotron 3.5 Lightning 30B A3B NVIDIA | 31.6B | 20 GB 19 GB + 0.4 GB cache | Speed not estimated for this architecture |
| Gemma 4 31B IT Google DeepMind | 31.3B | 22 GB 19 GB + 2.2 GB cache | 16–20 tokens/s Comfortable to read |
| GLM-4.7-Flash Zhipu AI (Z.ai) | 31.2B 5.2B active | 20 GB 19 GB + 0.4 GB cache | 36–46 tokens/s Fast |
| Qwen3 VL 30B A3B Instruct Alibaba (Qwen) | 31.1B 3.9B active | 20 GB 19 GB + 0.8 GB cache | 42–55 tokens/s Fast |
| Qwen3 VL 30B A3B Thinking Alibaba (Qwen) | 31.1B 3.9B active | 20 GB 19 GB + 0.8 GB cache | 42–55 tokens/s Fast |
| Qwen3-Coder 30B-A3B Instruct Alibaba (Qwen) | 30.5B 3.4B active | 20 GB 19 GB + 0.8 GB cache | 47–61 tokens/s Fast |
| Qwen3 30B A3B Instruct 2507 Alibaba (Qwen) | 30.5B 3.4B active | 20 GB 19 GB + 0.8 GB cache | 47–61 tokens/s Fast |
| Qwen3 30B A3B Thinking 2507 Alibaba (Qwen) | 30.5B 3.4B active | 20 GB 19 GB + 0.8 GB cache | 47–61 tokens/s Fast |
| Qwen3 30B A3B Alibaba (Qwen) | 30.5B 3.4B active | 20 GB 19 GB + 0.8 GB cache | 47–61 tokens/s Fast |
| Hy-MT2-30B-A3B Tencent (Hunyuan) | 30.1B 3.5B active | 20 GB 18 GB + 0.8 GB cache | 46–59 tokens/s Fast |
| Muse Glimmer 30B Meta | 29.8B | 19 GB 18 GB + 0.2 GB cache | 17–23 tokens/s Comfortable to read |
| Qwen3.8 27B Alibaba (Qwen) | 27.8B | 18 GB 17 GB + 0.5 GB cache | 18–24 tokens/s Comfortable to read |
| Qwen3.5 27B Alibaba (Qwen) | 27.8B | 18 GB 17 GB + 0.5 GB cache | 18–24 tokens/s Comfortable to read |
| Qwen3.6 27B Alibaba (Qwen) | 27.8B | 18 GB 17 GB + 0.5 GB cache | 18–24 tokens/s Comfortable to read |
| Gemma 3 27B IT Google DeepMind | 27.4B | 17 GB weights only | Speed not estimated (architecture not public) |
| Gemma 2 27B Google DeepMind | 27.2B | 17 GB weights only | Speed not estimated (architecture not public) |
| Gemma 4 26B A4B IT Google DeepMind | 25.8B | 17 GB 16 GB + 0.5 GB cache | Speed not estimated for this architecture |
| Voxtral Small 24B 2507 Mistral AI | 24.3B | 17 GB 15 GB + 1.3 GB cache | 20–26 tokens/s Comfortable to read |
| Uncensored Cognitivecomputations | 24.0B | 17 GB 15 GB + 1.3 GB cache | 21–27 tokens/s Comfortable to read |
| Mistral Small 3.1 24B Instruct Mistral AI | 24.0B | 17 GB 15 GB + 1.3 GB cache | 21–27 tokens/s Comfortable to read |
| LFM2.5-24B-A2B Liquid | 24.0B | 16 GB 15 GB + 0.2 GB cache | Speed not estimated for this architecture |
| Mistral Small 3 Mistral AI | 23.6B | 16 GB 14 GB + 1.3 GB cache | 21–27 tokens/s Comfortable to read |
| Cydonia 24B V4.1 Thedrummer | 23.6B | 16 GB 14 GB + 1.3 GB cache | 21–27 tokens/s Comfortable to read |
| Cydonia 24B V4.1 Thedrummer | 23.6B | 16 GB 14 GB + 1.3 GB cache | 21–27 tokens/s Comfortable to read |
| Safety GPT OSS 20B OpenAI | 21.5B 4.8B active | 14 GB 13 GB + 0.2 GB cache | 39–51 tokens/s Fast |
| GPT OSS 20B OpenAI | 20.9B 4.2B active | 14 GB 13 GB + 0.2 GB cache | 44–57 tokens/s Fast |
| Reka Flash 3 Rekaai | 20.9B | 15 GB 13 GB + 1.1 GB cache | 24–31 tokens/s Comfortable to read |
| Qwen3 14B Alibaba (Qwen) | 14.8B | 11 GB 9.0 GB + 1.3 GB cache | 32–41 tokens/s Fast |
| Phi 4 Microsoft | 14.7B | 11 GB 8.9 GB + 1.7 GB cache | 32–41 tokens/s Fast |
| Ministral 3 14B Mistral AI | 13.9B | 11 GB 8.5 GB + 1.3 GB cache | 34–44 tokens/s Fast |
| Gemma 3 12B IT Google DeepMind | 12.2B | 8.2 GB weights only | Speed not estimated (architecture not public) |
| Llama Guard 4 12B Meta | 12.0B | 8.1 GB weights only | Speed not estimated (architecture not public) |
| Qwen3.5 9B Alibaba (Qwen) | 9.7B | 6.9 GB 5.9 GB + 0.3 GB cache | 50–64 tokens/s Fast |
| Moondream 3.1 Moondream | 9.3B | 6.4 GB weights only | Speed not estimated (architecture not public) |
| Moondream 3 Preview Moondream | 9.3B | 6.4 GB weights only | Speed not estimated (architecture not public) |
| Ministral 3 8B Mistral AI | 8.9B | 7.3 GB 5.4 GB + 1.1 GB cache | 50–65 tokens/s Fast |
| Granite 4.2 8B IBM | 8.8B | 7.5 GB 5.3 GB + 1.3 GB cache | 50–64 tokens/s Fast |
| Qwen3 VL 8B Instruct Alibaba (Qwen) | 8.8B | 7.3 GB 5.3 GB + 1.2 GB cache | 50–65 tokens/s Fast |
| Qwen3 VL 8B Thinking Alibaba (Qwen) | 8.8B | 7.3 GB 5.3 GB + 1.2 GB cache | 50–65 tokens/s Fast |
| UI-TARS 7B Bytedance | 8.3B | 6.3 GB 5.0 GB + 0.5 GB cache | 56–73 tokens/s Fast |
| Qwen3 8B Alibaba (Qwen) | 8.2B | 7.0 GB 5.0 GB + 1.2 GB cache | 53–69 tokens/s Fast |
| Llama 3.1 8B Instruct fp8 Meta | 8.0B | 5.7 GB weights only | Speed not estimated (architecture not public) |
| Hy-MT2-7B Tencent (Hunyuan) | 8.0B | 6.7 GB 4.9 GB + 1.1 GB cache | 55–71 tokens/s Fast |
| Qwen2.5 7B Instruct Alibaba (Qwen) | 7.6B | 5.9 GB 4.6 GB + 0.5 GB cache | 60–78 tokens/s Fast |
| Gemma 3 4B IT Google DeepMind | 4.3B | 3.4 GB weights only | Speed not estimated (architecture not public) |
| Nemotron 3.5 Content Safety (free) NVIDIA | 4.3B | 3.7 GB 2.6 GB + 0.3 GB cache | 99–128 tokens/s Fast |
| Ministral 3 3B Mistral AI | 3.8B | 4.0 GB 2.3 GB + 0.9 GB cache | 98–127 tokens/s Fast |
| Llama 3.2 3B Instruct Meta | 3.2B | 2.7 GB weights only | Speed not estimated (architecture not public) |
| Granite 4.0 Micro IBM | 3.2B | 2.8 GB 1.9 GB + 0.1 GB cache | 129–167 tokens/s Fast |
| Hy-MT2-1.8B Tencent (Hunyuan) | 2.0B | 2.6 GB 1.2 GB + 0.5 GB cache | 159–205 tokens/s Fast |
| Hy-MT2-1.8B Tencent (Hunyuan) | 2.0B | 2.6 GB 1.2 GB + 0.5 GB cache | 159–205 tokens/s Fast |
| Llama 3.2 1B Instruct Meta | 1.2B | 1.5 GB weights only | Speed not estimated (architecture not public) |
| Runs, partly from system RAM (slower) (11) | |||
| Qwen3-Next 80B-A3B Instruct Alibaba (Qwen) | 81.3B 5.5B active | 51 GB 49 GB + 0.8 GB cache | 13–18 tokens/s Comfortable to read, 20 GB in system RAM |
| Qwen3-Next 80B-A3B (Thinking) Alibaba (Qwen) | 81.3B 5.5B active | 51 GB 49 GB + 0.8 GB cache | 13–18 tokens/s Comfortable to read, 20 GB in system RAM |
| Hunyuan A13B Instruct Tencent (Hunyuan) | 80.4B | 51 GB 49 GB + 1.1 GB cache | 19 GB in system RAM; Speed not estimated for this architecture |
| Qwen3 Coder Next Alibaba (Qwen) | 79.7B 3.9B active | 50 GB 48 GB + 0.8 GB cache | 15–21 tokens/s Comfortable to read, 19 GB in system RAM |
| Qwen2.5-VL 72B Instruct Alibaba (Qwen) | 73.4B | 48 GB 45 GB + 2.7 GB cache | 1.1–1.6 tokens/s Too slow for chat, 17 GB in system RAM |
| Qwen2.5 72B Instruct Alibaba (Qwen) | 72.7B | 48 GB 44 GB + 2.7 GB cache | 1.2–1.6 tokens/s Too slow for chat, 16 GB in system RAM |
| Magnum v4 72B Anthracite ORG | 72.7B | 48 GB 44 GB + 2.7 GB cache | 1.2–1.6 tokens/s Too slow for chat, 16 GB in system RAM |
| Hermes 3 70B Instruct Nousresearch | 70.6B | 46 GB 43 GB + 2.7 GB cache | 1.3–1.7 tokens/s Too slow for chat, 15 GB in system RAM |
| Hermes 3 70B Instruct Nousresearch | 70.6B | 46 GB 43 GB + 2.7 GB cache | 1.3–1.7 tokens/s Too slow for chat, 15 GB in system RAM |
| Llama-3.3-70B-Instruct Meta | 70.6B | 44 GB weights only | 12 GB in system RAM; Speed not estimated (architecture not public) |
| Llama-3.1-70B-Instruct Meta | 70.6B | 44 GB weights only | 12 GB in system RAM; Speed not estimated (architecture not public) |
74 larger models do not fit. Change the quantization, conversation length or system RAM in the tool.
Compared with similar cards
- NVIDIA GeForce RTX 4090 24 GB: 24 GB, 60 models fit entirely (-1); about 131% faster on models that fit both (1008 vs 608 GB/s, and Intel cards run llama.cpp less efficiently than NVIDIA's).
- NVIDIA GeForce RTX 5090 32 GB: same memory, so the same models fit; about 202% faster on models that fit both (1792 vs 608 GB/s, and Intel cards run llama.cpp less efficiently than NVIDIA's).
- AMD Radeon AI PRO R9700 32 GB: same memory, so the same models fit; about 35% faster on models that fit both (640 vs 608 GB/s, and Intel cards run llama.cpp less efficiently than AMD's).
Frequently asked questions
What is the largest AI model the Intel Arc Pro B70 32 GB can run?
At 4-bit quantization with an 8K-token conversation, the largest model in our directory that fits entirely in its 32 GB of video memory is Skyfall 36B V2 (25 GB). With 32 GB of dual-channel DDR5-5600 for the overflow, larger models such as Qwen3-Next 80B-A3B Instruct also run, more slowly, partly from system RAM.
Can the Intel Arc Pro B70 32 GB run GPT OSS 20B?
Yes. At 4-bit, GPT OSS 20B needs about 14 GB and fits entirely in the Intel Arc Pro B70 32 GB's memory at an estimated 44 to 57 tokens per second.
Can the Intel Arc Pro B70 32 GB run Qwen3 32B?
Yes. At 4-bit, Qwen3 32B needs about 23 GB and fits entirely in the Intel Arc Pro B70 32 GB's memory at an estimated 15 to 20 tokens per second.
Can the Intel Arc Pro B70 32 GB run OpenAI GPT 120B?
No. OpenAI GPT 120B needs about 72 GB at 4-bit, more than the Intel Arc Pro B70 32 GB can hold even with 32 GB of dual-channel DDR5-5600 for overflow.
These are estimates, not measurements: memory from each model's parameter count and published architecture, speed from memory bandwidth, calibrated on the AIPCs test bench. Drivers, software and prompt length move real results. How the estimate works.