diff --git a/src/banks/extensions/__init__.py b/src/banks/extensions/__init__.py
index 2f5bf74..bd78f6a 100644
--- a/src/banks/extensions/__init__.py
+++ b/src/banks/extensions/__init__.py
@@ -1,6 +1,3 @@
 # SPDX-FileCopyrightText: 2023-present Massimiliano Pippi <mpippi@gmail.com>
 #
 # SPDX-License-Identifier: MIT
-from banks.extensions.generate import GenerateExtension
-
-__all__ = ("GenerateExtension",)
diff --git a/src/banks/extensions/generate.py b/src/banks/extensions/generate.py
deleted file mode 100644
index 4cf08a3..0000000
--- a/src/banks/extensions/generate.py
+++ /dev/null
@@ -1,76 +0,0 @@
-# SPDX-FileCopyrightText: 2023-present Massimiliano Pippi <mpippi@gmail.com>
-#
-# SPDX-License-Identifier: MIT
-from typing import cast
-
-from deprecated import deprecated
-from jinja2 import nodes
-from jinja2.ext import Extension
-from litellm import acompletion, completion
-from litellm.types.utils import ModelResponse
-
-from banks.errors import CanaryWordError
-from banks.prompt import Prompt
-
-DEFAULT_MODEL = "gpt-3.5-turbo"
-SYSTEM_PROMPT = Prompt("{{canary_word}} You are a helpful assistant.")
-
-
-@deprecated(version="1.3.0", reason="This extension is deprecated, use {% completion %} instead.")
-class GenerateExtension(Extension):
-    # a set of names that trigger the extension.
-    tags = {"generate"}  # noqa
-
-    def parse(self, parser):
-        # We get the line number of the first token so that we can give
-        # that line number to the nodes we create by hand.
-        lineno = next(parser.stream).lineno
-
-        # The args passed to the extension:
-        # - the prompt text used to generate new text
-        # - (optional) the name of the model use to generate new text
-        args = [parser.parse_expression()]
-
-        # If there is a comma, the user provided the model name. If not, use
-        # None as the second parameter.
-        if parser.stream.skip_if("comma"):
-            args.append(parser.parse_expression())
-        else:
-            args.append(nodes.Const(None))
-
-        if parser.environment.is_async:
-            return nodes.Output([self.call_method("_agenerate", args)]).set_lineno(lineno)
-        return nodes.Output([self.call_method("_generate", args)]).set_lineno(lineno)
-
-    def _generate(self, text, model_name=DEFAULT_MODEL):
-        """
-        Helper callback.
-
-        To tweak the prompt used to generate content, change the variable `messages` .
-        """
-        messages = [
-            {"role": "system", "content": SYSTEM_PROMPT.text()},
-            {"role": "user", "content": text},
-        ]
-        response: ModelResponse = cast(ModelResponse, completion(model=model_name, messages=messages))
-        return self._get_content(response)
-
-    async def _agenerate(self, text, model_name=DEFAULT_MODEL):
-        """
-        Helper callback.
-
-        To tweak the prompt used to generate content, change the variable `messages` .
-        """
-        messages = [
-            {"role": "system", "content": SYSTEM_PROMPT.text()},
-            {"role": "user", "content": text},
-        ]
-        response: ModelResponse = cast(ModelResponse, await acompletion(model=model_name, messages=messages))
-        return self._get_content(response)
-
-    def _get_content(self, response: ModelResponse) -> str:
-        content = response["choices"][0]["message"]["content"]
-        if SYSTEM_PROMPT.canary_leaked(content):
-            msg = "The system prompt has leaked into the response, possible prompt injection!"
-            raise CanaryWordError(msg)
-        return content