[Feature] Supported deepseek_v4_fp8 quantization in rocm

4f799f11 · zhuwenwen · a7378418 · 4f799f11
Commit 4f799f11 authored Apr 29, 2026 by zhuwenwen
Hide whitespace changes
Inline Side-by-side

Showing with 1 addition and 0 deletions

vllm/platforms/rocm.py vllm/platforms/rocm.py +1 -0

No files found.
--- a/vllm/platforms/rocm.py
+++ b/vllm/platforms/rocm.py
@@ -418,6 +418,7 @@ class RocmPlatform(Platform):
        "torchao",
        "bitsandbytes",
        "modelopt_fp4",
+        "deepseek_v4_fp8",
    ]

    @classmethod