Llama 2 7B
on Intel Arc B580 12GB · llama.cpp
Summary
User reports Llama 2 7B Q4_0 prompt processing at 4,454 t/s on an Intel Arc B580 with llama.cpp's SYCL backend, ahead of the RTX 5060 Ti's 4,196 t/s on llama.cpp's scoreboard.
Prompt processing is 3.3x to 6.6x faster than llama.cpp master across Llama 2, Llama 3.1, Qwen3.5 and Gemma 4, with long chats 1.5x faster at 32K and 1.6x at 64K.
Short batches on K-quant models are up to 2.5x faster and still in progress; perplexity is unchanged. The changes are on the user's GitHub branches and submitted for upstreaming.