diff --git a/comfy/cli_args.py b/comfy/cli_args.py
index 79ecbd682..dfb9e3cef 100644
--- a/comfy/cli_args.py
+++ b/comfy/cli_args.py
@@ -130,6 +130,7 @@ parser.add_argument("--reserve-vram", type=float, default=None, help="Set the am
 
 parser.add_argument("--default-hashing-function", type=str, choices=['md5', 'sha1', 'sha256', 'sha512'], default='sha256', help="Allows you to choose the hash function to use for duplicate filename / contents comparison. Default is sha256.")
 
+parser.add_argument("--disable-fp8-compute", action="store_true", help="Prevent ComfyUI from activating fp8 compute in Nvidia cards that support it. Can prevent some issues with some models not suitable for fp8 compute.")
 parser.add_argument("--disable-smart-memory", action="store_true", help="Force ComfyUI to agressively offload to regular ram instead of keeping models in vram when it can.")
 parser.add_argument("--deterministic", action="store_true", help="Make pytorch use slower deterministic algorithms when it can. Note that this might not make images deterministic in all cases.")
 
diff --git a/comfy/model_management.py b/comfy/model_management.py
index 19e6c8dff..084093264 100644
--- a/comfy/model_management.py
+++ b/comfy/model_management.py
@@ -1211,6 +1211,8 @@ def should_use_bf16(device=None, model_params=0, prioritize_performance=True, ma
 def supports_fp8_compute(device=None):
     if not is_nvidia():
         return False
+    if args.disable_fp8_compute:
+        return False
 
     props = torch.cuda.get_device_properties(device)
     if props.major >= 9: