updated

Dockerfile

@@ -7,12 +7,15 @@ COPY ./requirements.txt /app/requirements.txt
 RUN pip3 install --no-cache-dir --upgrade pip
 RUN pip3 install --no-cache-dir wheel setuptools build
 RUN pip3 install --no-cache-dir --use-pep517 -r /app/requirements.txt
-
+ 
 # User
 RUN useradd -m -u 1000 user
 USER user
 ENV HOME /home/user
 ENV PATH $HOME/.local/bin:$PATH
+ENV TIKTOKEN_CACHE_DIR $HOME/.cache/tiktoken
+
+RUN mkdir -p $HOME/.cache/tiktoken
 
 WORKDIR $HOME
 RUN mkdir app

agent.py

@@ -5,6 +5,8 @@ from omegaconf import OmegaConf
 
 from vectara_agentic.agent import Agent
 from vectara_agentic.tools import VectaraToolFactory, ToolsFactory
+from vectara_agentic.agent_config import AgentConfig
+from vectara_agentic.types import ModelProvider, AgentType
 
 from dotenv import load_dotenv
 load_dotenv(override=True)
@@ -29,7 +31,6 @@ def create_assistant_tools(cfg):
         return tickers
 
     class QueryHMC(BaseModel):
-        query: str = Field(description="The user query.")
         ticker: Optional[str] = Field(
             default=None,
             description="The company ticker.",
@@ -80,8 +81,10 @@ def create_assistant_tools(cfg):
         n_sentences_before = 2, n_sentences_after = 2, lambda_val = 0.005,
         vectara_summarizer = summarizer,
         summary_num_results = 10,
+        max_tokens = 4096, max_response_chars = 8192,
         include_citations = True,
         verbose = True,
+        save_history = True,
     )
     tools_factory = ToolsFactory()
     return [ask_hmc] + [tools_factory.create_tool(get_company_info)]
@@ -95,6 +98,22 @@ def initialize_agent(_cfg, agent_progress_callback=None):
     - Note that 10Q reports exist for quarters 1, 2, 3 and for the 4th quarter there is a 10K report.
     - If the 'ask_hmc' tool does not return any results, check the year and ticker and try calling it again with the right values.
     """
+    agent_config = AgentConfig(
+        agent_type = os.getenv("VECTARA_AGENTIC_AGENT_TYPE", AgentType.OPENAI.value),
+        main_llm_provider = os.getenv("VECTARA_AGENTIC_MAIN_LLM_PROVIDER", ModelProvider.OPENAI.value),
+        main_llm_model_name = os.getenv("VECTARA_AGENTIC_MAIN_MODEL_NAME", ""),
+        tool_llm_provider = os.getenv("VECTARA_AGENTIC_TOOL_LLM_PROVIDER", ModelProvider.OPENAI.value),
+        tool_llm_model_name = os.getenv("VECTARA_AGENTIC_TOOL_MODEL_NAME", ""),
+        observer = os.getenv("VECTARA_AGENTIC_OBSERVER_TYPE", "NO_OBSERVER")
+    )
+    fallback_agent_config = AgentConfig(
+        agent_type = os.getenv("VECTARA_AGENTIC_FALLBACK_AGENT_TYPE", AgentType.OPENAI.value),
+        main_llm_provider = os.getenv("VECTARA_AGENTIC_FALLBACK_MAIN_LLM_PROVIDER", ModelProvider.OPENAI.value),
+        main_llm_model_name = os.getenv("VECTARA_AGENTIC_FALLBACK_MAIN_MODEL_NAME", ""),
+        tool_llm_provider = os.getenv("VECTARA_AGENTIC_FALLBACK_TOOL_LLM_PROVIDER", ModelProvider.OPENAI.value),
+        tool_llm_model_name = os.getenv("VECTARA_AGENTIC_FALLBACK_TOOL_MODEL_NAME", ""),
+        observer = os.getenv("VECTARA_AGENTIC_OBSERVER_TYPE", "NO_OBSERVER")
+    )
 
     agent = Agent(
         tools=create_assistant_tools(_cfg),
@@ -102,6 +121,8 @@ def initialize_agent(_cfg, agent_progress_callback=None):
         custom_instructions=bot_instructions,
         agent_progress_callback=agent_progress_callback,
         verbose=True,
+        agent_config=agent_config,
+        fallback_agent_config=fallback_agent_config,
     )
     agent.report()
     return agent

requirements.txt

@@ -1,9 +1,9 @@
 omegaconf==2.3.0
 python-dotenv==1.0.1
-streamlit==1.43.2
+streamlit==1.45.0
 streamlit_feedback==0.1.3
 uuid==1.30
 langdetect==1.0.9
 langcodes==3.4.0
-vectara-agentic==0.2.9
-torch==2.6.0
+vectara-agentic==0.2.15
+torch==2.6.0

st_app.py

@@ -139,7 +139,7 @@ async def launch_bot():
     if st.session_state.prompt:
         with st.chat_message("assistant", avatar='🤖'):
             st.session_state.status = st.status('Processing...', expanded=False)
-            response = st.session_state.agent.chat(st.session_state.prompt)
+            response = await st.session_state.agent.achat(st.session_state.prompt)
             res = escape_dollars_outside_latex(response.response)
             message = {"role": "assistant", "content": res, "avatar": '🤖'}
             st.session_state.messages.append(message)