llama.cpp
101428bb
- metal : rename the FA dequant_f16 identifiers to kv_f16
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 day ago
metal : rename the FA dequant_f16 identifiers to kv_f16 Assisted-by: pi:llama.cpp/Qwen3.8-27B
References
#27390 - metal : dequantize quantized KV to F16 before flash attention
Author
ggerganov
Parents
275768fb
Loading