diff --git a/.env.docker b/.env.docker
index 1052f00..10b923a 100644
--- a/.env.docker
+++ b/.env.docker
@@ -1,5 +1,42 @@
+# Copyright 2024 Anirban Basu
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 # Application-specific Gradio configuration
-GRADIO__SERVER_HOST="0.0.0.0"
+GRADIO__SERVER_HOST = "0.0.0.0"
 
 # Ollama URL assuming that it is on the Docker host
-LLM__OLLAMA_URL="http://host.docker.internal:11434"
+LLM__PROVIDER = "Ollama"
+
+# Anthropic
+LLM__ANTHROPIC_API_KEY = "your-anthropic-api-key"
+LLM__ANTHROPIC_MODEL = "claude-3-opus-20240229"
+
+# OpenAI
+LLM__OPENAI_API_KEY = "your-openai-api-key"
+LLM__OPENAI_MODEL = "gpt-4o-mini"
+
+# Cohere
+LLM__COHERE_API_KEY = "your-cohere-api-key"
+LLM__COHERE_MODEL = "command-r-plus"
+
+# Groq
+LLM__GROQ_API_KEY = "the-groq-api-key"
+LLM__GROQ_MODEL = "llama-3.1-70b-versatile"
+
+# Ollama
+LLM__OLLAMA_URL = "http://host.docker.internal:11434"
+# The model must be available in the Ollama installation.
+LLM__OLLAMA_MODEL = "mistral-nemo"
+
+LLM__TEMPERATURE = "0.0"
diff --git a/src/coder_agent.py b/src/coder_agent.py
index 1099a9c..d0a4e80 100644
--- a/src/coder_agent.py
+++ b/src/coder_agent.py
@@ -174,7 +174,10 @@ def solve(self, state: AgentState) -> dict:
             ]
         else:
             inputs[constants.AGENT_STATE__KEY_EXAMPLES] = constants.EMPTY_STRING
-        response = self.pydantic_to_ai_message(self._solver_agent.invoke(inputs))
+
+        llm_response = self._solver_agent.invoke(inputs)
+        ic(llm_response)
+        response = self.pydantic_to_ai_message(llm_response)
         ic(response)
         return (
             {
@@ -238,6 +241,11 @@ def evaluate(self, state: AgentState) -> dict:
                     )
                 ]
             }
+        num_test_cases = len(test_cases) if test_cases is not None else 0
+        if num_test_cases == 0:
+            return {
+                constants.AGENT_STATE__KEY_STATUS: constants.AGENT_NODE__EVALUATE_STATUS_NO_TEST_CASES
+            }
         try:
             # Extract the code from the tool call.
             code: str = json_dict[constants.PYDANTIC_MODEL__CODE_OUTPUT__CODE]
@@ -255,11 +263,6 @@ def evaluate(self, state: AgentState) -> dict:
                     )
                 ]
             }
-        num_test_cases = len(test_cases) if test_cases is not None else 0
-        if num_test_cases == 0:
-            return {
-                constants.AGENT_STATE__KEY_STATUS: constants.AGENT_NODE__EVALUATE_STATUS_NO_TEST_CASES
-            }
         succeeded = 0
         test_results = []
         # TODO: Enable parallel execution of test cases.
diff --git a/src/constants.py b/src/constants.py
index 6b559b2..1e94de3 100644
--- a/src/constants.py
+++ b/src/constants.py
@@ -130,6 +130,7 @@
 First, output a `reasoning` through the problem and conceptualise a solution. Whenever possible, add a time and a space complexity analysis for your solution.
 Then, output a `pseudocode` in Pascal to implement your concept solution.
 Finally, a well-documented working Python 3 `code` for your solution. Do not use external libraries. Your code must be able to accept inputs from `sys.stdin` and write the final output to `sys.stdout` (or, to `sys.stderr` in case of errors).
+Please format your response as JSON, using `reasoning`, `pseudocode`, and `code` as attributes.
 
 Optional examples of similar problems and solutions (may not be in Python):
 {examples}
@@ -145,6 +146,7 @@
 First, output a `reasoning` through the problem and conceptualise a solution. Whenever possible, add a time and a space complexity analysis for your solution.
 Then, output a `pseudocode` in Pascal to implement your concept solution.
 Finally, a well-documented working Python 3 `code` for your solution. Do not use external libraries. Your code must be able to accept inputs from `sys.stdin` and write the final output to `sys.stdout` (or, to `sys.stderr` in case of errors).
+Please format your response as JSON, using `reasoning`, `pseudocode`, and `code` as attributes.
 
 Optional examples of similar problems and solutions (may not be in Python):
 {examples}
diff --git a/src/webapp.py b/src/webapp.py
index 6648f03..ff83532 100644
--- a/src/webapp.py
+++ b/src/webapp.py
@@ -70,6 +70,8 @@ def __init__(self):
             default_value=constants.ENV_VAR_VALUE__LLM_PROVIDER,
         )
         if self._llm_provider == "Ollama":
+            # ChatOllama does not support structured outputs: https://python.langchain.com/v0.2/docs/integrations/chat/ollama/
+            # Yet, the API docs seems to suggest that it does: https://api.python.langchain.com/en/latest/chat_models/langchain_community.chat_models.ollama.ChatOllama.html#langchain_community.chat_models.ollama.ChatOllama.with_structured_output
             self._llm = ChatOllama(
                 base_url=parse_env(
                     constants.ENV_VAR_NAME__LLM_OLLAMA_URL,