Activity
[CPU]add comment for rotate_kv_cache
[CPU]add comment for rotate_kv_cache
[CPU]fix reference kernel of quant by channel
[CPU]fix reference kernel of quant by channel
Force push
[CPU]fix reference kernel of quant by channel
[CPU]fix reference kernel of quant by channel
[CPU]attn_acc_value_block with avx2 support
[CPU]attn_acc_value_block with avx2 support
[CPU]ENABLE value cache U4/U8 by-channel quant with avx512
[CPU]ENABLE value cache U4/U8 by-channel quant with avx512
Deleted branch
Bump actions/download-artifact from 4.1.9 to 4.2.1
Bump actions/download-artifact from 4.1.9 to 4.2.1
5 days ago
Deleted branch
Bump actions/upload-artifact from 4.4.3 to 4.6.2
Bump actions/upload-artifact from 4.4.3 to 4.6.2
5 days ago
Bump actions/download-artifact from 4.1.8 to 4.2.0
Bump actions/download-artifact from 4.1.8 to 4.2.0
Force push
6 days ago
[CPU]enable u4 from make_pa_executor
[CPU]enable u4 from make_pa_executor
Force push
[CPU]enable u4 from make_pa_executor
[CPU]enable u4 from make_pa_executor
Deleted branch
Bump actions/download-artifact from 4.1.8 to 4.2.0
Bump actions/download-artifact from 4.1.8 to 4.2.0
6 days ago
[CPU]vectorize dequant_u4
[CPU]vectorize dequant_u4
[CPU]Fix by_chhanel quant for avx2
[CPU]Fix by_chhanel quant for avx2
Deleted branch
Bump actions/download-artifact from 4.1.8 to 4.1.9
Bump actions/download-artifact from 4.1.8 to 4.1.9
8 days ago
Bump actions/setup-node from 4.2.0 to 4.3.0
Bump actions/setup-node from 4.2.0 to 4.3.0
Deleted branch
Bump actions/upload-artifact from 4.4.3 to 4.6.1
Bump actions/upload-artifact from 4.4.3 to 4.6.1
8 days ago
[CPU]Support u4 by-dim quant for key cache
[CPU]Support u4 by-dim quant for key cache
[CPU]Vectorize dot_product by_channel
[CPU]Vectorize dot_product by_channel
[CPU]init support for u4 by channel quantization
[CPU]init support for u4 by channel quantization