Intel Arc Pro B65 for local LLMs
Written by Jakub Rusinowski · Last updated
With 32 GB of GDDR6 at 608 GB/s, the Intel Arc Pro B65 runs 122 catalogued models at Q4_K_M with 8K context. The largest that fits is K2 Horizon MoVA 36B-A4B (~25 GB), and the top pick is K2 Horizon MoVA 36B-A4B at 43–90 tok/s.
The B65 keeps the B70's 32 GB and 608 GB/s but drops to 20 Xe cores of compute. For inference, which is bandwidth-bound far more than compute-bound, that matters less than the core count suggests.
Models that run on the Intel Arc Pro B65
Q4_K_M, 8K context, 32 GB usable. Ranked by quality and speed.
| Model | Memory · marker = 32 GB | VRAM | Speed |
|---|---|---|---|
| K2 Horizon MoVA 36B-A4B K2 Horizon | 25 GB | 43–90 tok/s | |
| Qwen 3.6 35B-A3B Qwen 3.6 | 22.2 GB | 68–141 tok/s | |
| Nex-N2.5 mini Nex-N2.5 | 21.9 GB | 46–96 tok/s | |
| Nex-N2 mini Nex-N2 | 21.9 GB | 46–96 tok/s | |
| Qwen 3.5 35B-A3B Qwen 3.5 | 21.9 GB | 46–96 tok/s | |
| Laguna XS 2.1 33B-A3B Poolside Laguna XS 2.1 | 20.7 GB | 46–96 tok/s | |
| Qwen 3 32B Qwen 3 | 22.8 GB | 10–21 tok/s | |
| Aya Expanse 32B Aya Expanse | 21.6 GB | 11–22 tok/s | |
| EXAONE 3.5 32B EXAONE 3.5 | 22.3 GB | 10–21 tok/s | |
| Cogito v1 32B Cogito v1 | 22.3 GB | 10–21 tok/s | |
| Granite 4.0 Small-H 32B-A9B IBM Granite 4.0 | 20.1 GB | 27–56 tok/s | |
| Olmo 3 32B Olmo 3 | 20.1 GB | 10–22 tok/s |
Buy it or rent the same memory
Buy the card, or rent a GPU with the same memory by the hour to try models first.
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
Speed vs other GPUs
Llama 3.1 8B, Q4_K_M. Estimated ranges. How this is calculated
Specifications
Specs last updated 2026-09-19.
- Memory
- 32 GB GDDR6
- Memory bandwidth
- 608 GB/s
- Memory bus
- 256-bit
- Architecture
- Xe2 Battlemage BMG-G31
- Series
- Arc Pro
- Board power
- 200 W
- Release year
- 2026
- Compute backends
- VULKAN, SYCL
- Usable for models
- 32 GB
Similar GPUs
Frequently asked questions
Can the Intel Arc Pro B65 run local LLMs?
Yes. With 32 GB (32 GB usable by a model) it runs 122 of the catalogued models at Q4_K_M with 8K context; the largest is K2 Horizon MoVA 36B-A4B, needing about 25 GB.
How fast is the Intel Arc Pro B65 for AI inference?
It is estimated to run Llama 3.1 8B at 35–72 tok/s at Q4_K_M. Llama 3.3 70B does not fit: it needs about 44 GB against 32 GB usable. These are modelled estimates from memory bandwidth, not measurements; the methodology page shows the formula.
What LLMs can I run on 32 GB?
Among the best that fit: K2 Horizon MoVA 36B-A4B, Qwen 3.6 35B-A3B, Nex-N2.5 mini, Nex-N2 mini, Qwen 3.5 35B-A3B.