diff --git a/backends/exllamav3/model.py b/backends/exllamav3/model.py
index 9e69074..24c08f3 100644
--- a/backends/exllamav3/model.py
+++ b/backends/exllamav3/model.py
@@ -99,7 +99,7 @@ class ExllamaV3Container(BaseModelContainer):
         self = cls()
 
         # Make sure ExllamaV3 is up to date
-        check_package_version("exllamav3", "0.0.3")
+        check_package_version("exllamav3", "0.0.4")
 
         logger.warning(
             "ExllamaV3 is currently in an alpha state. "
diff --git a/pyproject.toml b/pyproject.toml
index ca69823..6a3a6ca 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -78,14 +78,14 @@ cu121 = [
     "exllamav2 @ https://github.com/turboderp-org/exllamav2/releases/download/v0.3.1/exllamav2-0.3.1+cu128.torch2.7.0-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
 
     # Exl3
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp313-cp313-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.13'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp312-cp312-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.12'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp313-cp313-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.13'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp312-cp312-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.12'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'",
-    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.3/exllamav3-0.0.3+cu128.torch2.7.0-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp313-cp313-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.13'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp312-cp312-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.12'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp311-cp311-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.11'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp310-cp310-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.10'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp313-cp313-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.13'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp312-cp312-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.12'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp311-cp311-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.11'",
+    "exllamav3 @ https://github.com/turboderp-org/exllamav3/releases/download/v0.0.4/exllamav3-0.0.4+cu128.torch2.7.0-cp310-cp310-linux_x86_64.whl ; platform_system == 'Linux' and platform_machine == 'x86_64' and python_version == '3.10'",
 
     # Windows FA2 from https://github.com/kingbri1/flash-attention/releases
     "flash_attn @ https://github.com/kingbri1/flash-attention/releases/download/v2.7.4.post1/flash_attn-2.7.4.post1+cu128torch2.7.0cxx11abiFALSE-cp313-cp313-win_amd64.whl ; platform_system == 'Windows' and python_version == '3.13'",