llama.cpp on the Intel Arc Pro B60 24GB
On the Intel Arc Pro B60 24GB, llama.cpp runs Qwen3.8 27B at 4-bit at a median 18.8 tokens per second for generation, from 2 plain runs (512 to 150K context).
Speed by model on the Intel Arc Pro B60 24GB
Plain runs only: one device, one request, no speculative decoding, the whole model in memory. Each row is one model size at one bit class, with the other engines' plain runs of the same model on this card beside it. Context length and build still differ between runs, and each side shows its own.
| Model | llama.cpp | Other engines |
|---|---|---|
| Qwen3.8 27B 4-bit | 18.8 t/smedian of 2, 512 to 150K context | no other engine reported |
| Qwen3.5 35B · 3B active 4-bit | 8.3 t/sone report, context not stated | no other engine reported |
6 llama.cpp reports on the Intel Arc Pro B60 24GB
Newest first. A run on several cards, with speculative decoding, with part of the model in system RAM or with batched requests says so, since each describes a different setup.
- Qwen3.8 27B · Q4_K_M53.3 t/s(speculative decoding)
- Nemotron 3.5 Lightning 30B · 3B active · Q4_091.9 t/s(2 cards, speculative decoding)
- Qwen3.8 27B · Q4_K_M15.3 t/s
- Qwen3.8 27B · Q4_K_M24.0 t/s(2 cards, speculative decoding)
- Qwen3.5 35B · 3B active · Q4_K_XL8.3 t/s
- Qwen3.8 27B · Q4_K22.2 t/s
Frequently asked
How fast is llama.cpp on the Intel Arc Pro B60 24GB?
On the Intel Arc Pro B60 24GB, llama.cpp runs Qwen3.8 27B at 4-bit at a median 18.8 tokens per second for generation, from 2 plain runs (512 to 150K context).