Update for Transformers AWQ support

2025-02-03 00:08:39 +08:00
parent ea6d26bf32
commit e2489ed4e3
15 changed files with 94055 additions and 58 deletions
--- a/quant_config.json
+++ b/quant_config.json
@@ -0,0 +1,6 @@
+{
+    "zero_point": true,
+    "q_group_size": 128,
+    "w_bit": 4,
+    "version": "GEMM"
+}