Switched to llama3.1-8B-instruct for summarization and qwen2.5-coder-8B-instruct for agent

configuration.py

@@ -11,20 +11,30 @@ DEFAULT_GITHUB_PROFILE = "https://github.com/gperdrizet"
 #     max_tokens=8000
 # )
 
+# Will be used for single shot summarization with no-frills prompting
+# (e.g. job call extraction). It needs to output JSON formatted text,
+# but this task does not require any complex reasoning or planning.
 SUMMARIZER_MODEL = OpenAI(
-    base_url="https://gperdrizet--deepseek-R1-qwen-32B-serve.modal.run/v1",
+    base_url="https://gperdrizet--llama-3-1-8b-instruct-serve.modal.run/v1",
     api_key=os.environ['MODAL_TOKEN_SECRET']
 )
 
+# Will be used for resume resume writing agent via HuggingFace smolagents
+# Including selection of relevant projects from GitHub profile
+#
+# Notes:
+# - DeepSeek-R1-Distill-Qwen-32B does not seem to work well with smolagents,
+#   has trouble correctly formatting responses as code.
+
 AGENT_MODEL = OpenAIServerModel(
-    model_id="deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", # Same as HF model string
-    api_base="https://gperdrizet--deepseek-R1-qwen-32B-serve.modal.run/v1",
+    model_id="Qwen/Qwen2.5-Coder-14B-Instruct", # Same as HF model string
+    api_base="https://gperdrizet--qwen2-5-coder-14b-instruct-serve.modal.run/v1",
     api_key=os.environ["MODAL_TOKEN_SECRET"],
 )
 
 INSTRUCTIONS = """
 You are an AI agent responsible for writing a resume based on the provided context. Your task is to generate a well-structured and professional resume that highlights the user's skills, experiences, and achievements.
-You will receive Two pieces of JSON structured context: a job call and a LinkedIn profile. 
+You will receive two pieces of JSON structured context: a job call and a LinkedIn profile. 
 
 LINKEDIN PROFILE EXAMPLE