mirror of
				https://github.com/ggml-org/llama.cpp.git
				synced 2025-10-31 08:51:55 +00:00 
			
		
		
		
	graph : make FA compatible with MLA + add initial Metal kernels (#12953)
* graph : make mla compatible with FA * metal : add exp FA kernels for DeepSeek models ggml-ci * llama : minor naming updates ggml-ci * ggml : disable FA for DS head sizes * tests : add FA tests for MLA shapes ggml-ci
This commit is contained in:
		| @@ -9261,6 +9261,7 @@ static bool ggml_backend_vk_device_supports_op(ggml_backend_dev_t dev, const ggm | ||||
|                 case 112: | ||||
|                 case 128: | ||||
|                 case 256: | ||||
|                 case 575: // DeepSeek MLA | ||||
|                     break; | ||||
|                 default: | ||||
|                     return false; | ||||
|   | ||||
		Reference in New Issue
	
	Block a user
	 Georgi Gerganov
					Georgi Gerganov