biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

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

Browse Source 
Signed-off-by: feng <fengli1702@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
This commit is contained in:
Daifeng Li
2025-08-23 10:53:21 +08:00
committed by GitHub
parent f6818a92cb
commit fa78de9dc3
8 changed files with 451 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/platforms/rocm.py
View File

@@ -171,7 +171,7 @@ class RocmPlatform(Platform):
supported_quantization: list[str] = [
"awq", "gptq", "fp8", "compressed-tensors", "fbgemm_fp8", "gguf",
"quark", "ptpc_fp8", "mxfp4"
"quark", "ptpc_fp8", "mxfp4", "petit_nvfp4"
]
@classmethod