Quantization: support FP4 quantized models on AMD CDNA2/CDNA3 GPUs (#22527)

Signed-off-by: feng <fengli1702@gmail.com> Signed-off-by: Michael Goin <mgoin64@gmail.com> Co-authored-by: Michael Goin <mgoin64@gmail.com>
2025-08-23 10:53:21 +08:00
parent f6818a92cb
commit fa78de9dc3
8 changed files with 451 additions and 5 deletions
--- a/vllm/model_executor/layers/linear.py
+++ b/vllm/model_executor/layers/linear.py
@@ -52,6 +52,7 @@ WEIGHT_LOADER_V2_SUPPORTED = [
    "HQQMarlinMethod",
    "QuarkLinearMethod",
    "ModelOptNvFp4LinearMethod",
+    "PetitNvFp4LinearMethod",
 ]