Update for Transformers AWQ support

2025-01-27 22:37:27 +08:00
parent 7b550f7fff
commit 79334b97df
14 changed files with 93970 additions and 60 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"
+}