Update for Transformers AWQ support

2025-01-26 19:00:16 +08:00
parent d03c2d995e
commit edfed77a30
15 changed files with 94148 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"
+}