Topic: sdpa

1 stories found

Today

releases48

ggml/llama.cpp releases: b10255

The ggml/llama.cpp project updated its SYCL oneDNN SDPA to support Q4_0-Q8_0 and FP32 KV caches, extending the handling of non-FP16 key-value pairs. This update is significant as it enhances the flexibility and performance of the model in various computational scenarios.

github.com↗

🌿 That's all for now. Come back tomorrow.

1 of 1 items shown. Sources: 76 days indexed.