Merge remote-tracking branch 'origin/v0.7.2-dev' into v0.7.2_zero_overhead

ca4ec0ce · lizhigong · 0be169ad · ae0ed592 · 0be169ad · 0be169ad
Commit ca4ec0ce authored Mar 25, 2025 by lizhigong
20 changed files
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_3584_3584_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_3584_3584_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_37888_3584_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_37888_3584_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_11008_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_11008_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_14336_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_14336_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_4096_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4096_4096_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4608_3584_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_4608_3584_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_13824_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_13824_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_2560_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_2560_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_5120_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_5120_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_6912_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_6912_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_8192_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_5120_8192_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_6144_4096_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_6144_4096_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_7168_8192_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_7168_8192_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_7680_5120_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_7680_5120_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_1024_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_1024_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_14336_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_14336_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_2048_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_2048_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_3584_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_3584_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_4096_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_4096_K100_AI.json
--- a/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_7168_K100_AI.json
+++ b/vllm/model_executor/layers/quantization/configs/w8a8/W8A8_8192_7168_K100_AI.json