You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

傅剑寒 121d7ad629 [Inference] Delete duplicated copy_vector (#5716)		7 months ago
..
attention	[Inference/Kernel] Optimize paged attention: Refactor key cache layout (#5643)	7 months ago
utils	[Inference] Delete duplicated copy_vector (#5716)	7 months ago
activation_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
context_kv_cache_memcpy_kernel.cu	[Inference/Feat] Add kvcache quant support for fused_rotary_embedding_cache_copy (#5680)	7 months ago
convert_fp8_kernel.cu	[Inference/Feat] Add convert_fp8 op for fp8 test in the future (#5706)	7 months ago
decode_kv_cache_memcpy_kernel.cu	[Inference] Delete duplicated copy_vector (#5716)	7 months ago
flash_decoding_attention_kernel.cu	add paged-attetionv2: support seq length split across thread block (#5707)	7 months ago
fused_rotary_emb_and_cache_kernel.cu	[Inference] Delete duplicated copy_vector (#5716)	7 months ago
get_cos_and_sin_kernel.cu	[Inference] Delete duplicated copy_vector (#5716)	7 months ago
layer_norm_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
moe_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_adam_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_apply.cuh	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_l2norm_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_lamb_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_scale_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
multi_tensor_sgd_kernel.cu	[Inference/Refactor] Refactor compilation mechanism and unified multi hw (#5613)	7 months ago
rms_layernorm_kernel.cu	[Inference] Remove unnecessary float4_ and rename float8_ to float8 (#5679)	7 months ago
scaled_masked_softmax_kernel.cu	[Inference] Delete duplicated copy_vector (#5716)	7 months ago
scaled_upper_triang_masked_softmax_kernel.cu	[Inference] Delete duplicated copy_vector (#5716)	7 months ago