Update for Transformers AWQ support

2025-02-15 00:56:32 +08:00
parent 9df68ab171
commit 4e56254ecb
15 changed files with 93932 additions and 55 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"
+}