GPUs & Macs › Intel Arc Pro B70
Intel Arc Pro B70: local LLM speed
Llama 2 7B (Q4_0) generates 114 tokens/s on the Intel Arc Pro B70 (Vulkan), and processes prompts at 3,151 tokens/s.
Measured results
| Model | Generate tok/s | Prompt tok/s | Quant | Backend | Test | Source |
|---|---|---|---|---|---|---|
| Llama 2 7B | 114 | 3,151 | Q4_0 | Vulkan FA on | tg128 | llama.cpp #10879 details |
| Llama 2 7B | 112 | 3,379 | Q4_0 | Vulkan FA off | tg128 | llama.cpp #10879 details |
What fits on the Intel Arc Pro B70
Estimates for this chip are scaled from a measured run: Llama 2 7B Q4_0 at 114 tok/s (Vulkan, source). Fit assumes the whole model in GPU memory with ~8K context.
Memory size unknown for this device.
Similar hardware
Intel Arc A750 · Intel Arc B570 · Intel Arc B580 · Intel Arc A770