aiagents4pharma 1.40.0__tar.gz → 1.41.0__tar.gz

{aiagents4pharma-1.40.0 → aiagents4pharma-1.41.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aiagents4pharma
-Version: 1.40.0
+Version: 1.41.0
 Summary: AI Agents for drug discovery, drug development, and other pharmaceutical R&D.
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License
@@ -20,6 +20,7 @@ Requires-Dist: langchain==0.3.7
 Requires-Dist: langchain-community==0.3.5
 Requires-Dist: langchain-core==0.3.40
 Requires-Dist: langchain-experimental==0.3.3
+Requires-Dist: langchain-milvus==0.2.0
 Requires-Dist: langchain-nvidia-ai-endpoints==0.3.9
 Requires-Dist: langchain-openai==0.2.5
 Requires-Dist: langchain_ollama==0.2.3
@@ -37,6 +38,7 @@ Requires-Dist: pypdf==5.2.0
 Requires-Dist: pytest==8.3.3
 Requires-Dist: pytest-asyncio==0.25.2
 Requires-Dist: pyzotero==1.6.9
+Requires-Dist: pymilvus==2.5.11
 Requires-Dist: streamlit==1.39.0
 Requires-Dist: sentence_transformers==3.3.1
 Requires-Dist: tabulate==0.9.0
@@ -111,103 +113,12 @@ Our toolkit currently consists of the following agents:
 
 _We now have all the agents available on Docker Hub._
 
-##### **To run Talk2AIAgents4Pharma / Talk2KnowledgeGraphs**
-
-If your machine has NVIDIA GPU(s), please install the following this:
-- [nvidia-cuda-toolkit](https://developer.nvidia.com/cuda-toolkit)
-- [nvidia-container-toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/1.17.8/install-guide.html) (required for GPU support with Docker; enables containers to access NVIDIA GPUs for accelerated computing). After installing `nvidia-container-toolkit`, please restart Docker to ensure GPU support is enabled.
-
-##### **1. Download docker-compose.yml, .env.example and startup.sh from GitHub**
-
-###### Talk2Agents4Pharma
-
-```sh
-mkdir talk2aiagents4pharma && cd talk2aiagents4pharma && wget https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/docker-compose.yml https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/.env.example https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/startup.sh
-```
-
-###### Talk2KnowledgeGraphs
-
-```sh
-mkdir talk2knowledgegraphs && cd talk2knowledgegraphs && wget https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/docker-compose.yml https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/.env.example https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/startup.sh
-```
-
-##### **2. Setup environment variables**
-
-Copy and configure your `.env` file:
-
-```sh
-cp .env.example .env
-```
-
-Then edit `.env` and add your API keys:
-
-```env
-OPENAI_API_KEY=...                  # Required for both agents
-NVIDIA_API_KEY=...                  # Required for both agents
-LANGCHAIN_TRACING_V2=true           # Optional for both agents
-LANGCHAIN_API_KEY=...               # Optional for both agents
-```
-
-[Additional Notes for Windows Users](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#notes-for-windows-users)
-
-##### **3. Start the application**
-
-```sh
-chmod +x startup.sh
-./startup.sh        # Add --cpu flag to force CPU mode if needed
-```
-
-[More about startup script](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#about-startupsh)
-
-##### **To Run Talk2Biomodels / Talk2Scholars**
-
-###### Talk2Biomodels
-
-```docker
-docker run -d \
---name talk2biomodels \
--e OPENAI_API_KEY=<your_openai_api_key> \
--e NVIDIA_API_KEY=<your_nvidia_api_key> \
--p 8501:8501 \
-virtualpatientengine/talk2biomodels
-```
-
-###### Talk2Scholars
-
-```docker
-docker run -d \
---name talk2scholars \
--e OPENAI_API_KEY=<your_openai_api_key> \
--e ZOTERO_API_KEY=<your_zotero_api_key> \
--e ZOTERO_USER_ID=<your_zotero_user_id> \
--e NVIDIA_API_KEY=<your_nvidia_api_key> \
--p 8501:8501 \
-virtualpatientengine/talk2scholars
-```
-
-##### **4. Access the Web UI**
-
-Once started, the agent is available at:
-
-```
-http://localhost:8501
-```
-
-To use **Talk2AIAgents4Pharma** or **Talk2KnowledgeGraphs**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-To use **Talk2BioModels** or **Talk2Scholars**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-Only for **Talk2Scholars**, you also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(For all other agents, the Zotero key is not required.)_
-
-If you are using docker on Windows, please follow these [Windows Setup Notes](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#notes-for-windows-users).
-
-**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
-
-[More on running multiple agents simultaneously](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#to-run-multiple-agents-simultaneously)
-
-📝 By default, `talk2knowledgegraphs` includes a small subset of the PrimeKG knowledge graph, allowing users to start interacting with it out of the box.  
-To switch to a different knowledge graph or use your own, refer to the [deployment guide](https://virtualpatientengine.github.io/AIAgents4Pharma/talk2knowledgegraphs/deployment/).
+Choose your agent below for detailed Docker instructions:
 
+- [Talk2AIAgents4Pharma](aiagents4pharma/talk2aiagents4pharma/install.md)
+- [Talk2KnowledgeGraphs](aiagents4pharma/talk2knowledgegraphs/install.md)
+- [Talk2BioModels](aiagents4pharma/talk2biomodels/install.md)
+- [Talk2Scholars](aiagents4pharma/talk2scholars/install.md)
 
 #### Option 2: git (for developers and contributors)
 
@@ -238,23 +149,6 @@ export LANGCHAIN_TRACING_V2=true    # Optional for all agents
 export LANGCHAIN_API_KEY=...        # Optional for all agents
 ```
 
-To use **Talk2AIAgents4Pharma**, **Talk2BioModels**, **Talk2KnowledgeGraphs**, or **Talk2Scholars**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-Only for **Talk2Scholars**, you also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(For all other agents, the Zotero key is not required.)_
-
-To use **Talk2Scholars**, you must have **FAISS** installed through **Conda**. Follow installation instructions for your OS [here](https://github.com/VirtualPatientEngine/AIAgents4Pharma/tree/main/aiagents4pharma/talk2scholars/install.md).
-
-Additionally on **Windows**, the `pcst_fast 1.0.10` library requires **Microsoft Visual C++ 14.0 or greater**.  
-You can download the **Microsoft C++ Build Tools** [here](https://visualstudio.microsoft.com/visual-cpp-build-tools/).
-
-**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
-
-_Please note that this will create a new tracing project in your Langsmith
-account with the name `T2X-xxxx`, where `X` can be `AA4P` (Main Agent),
-`B` (Biomodels), `S` (Scholars), `KG` (KnowledgeGraphs), or `C` (Cells).
-If you skip the previous step, it will default to the name `default`.
-`xxxx` will be the 4-digit ID created for the session._
-
 4. **Launch the app:**
 
 ```sh
@@ -269,10 +163,28 @@ _Replace `<agent>` with the agent name you are interested to launch:_
 - `talk2scholars`
 - `talk2cells`
 
+If your machine has NVIDIA GPU(s), please install the following this:
+
+- [nvidia-cuda-toolkit](https://developer.nvidia.com/cuda-toolkit)
+- [nvidia-container-toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/1.17.8/install-guide.html) (required for GPU support with Docker; enables containers to access NVIDIA GPUs for accelerated computing). After installing `nvidia-container-toolkit`, please restart Docker to ensure GPU support is enabled.
+
+To use the **Agents**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
+
+**Talk2Scholars** requires Milvus to be set up as the vector database — install Milvus depending on your setup by following the official instructions for [CPU](https://milvus.io/docs/install_standalone-docker-compose.md) or [GPU](https://milvus.io/docs/install_standalone-docker-compose-gpu.md). You will also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(The Zotero key is only required for Talk2Scholars; all other agents do not need it.)_
+
+> Additionally on **Windows**, the `pcst_fast 1.0.10` library requires **Microsoft Visual C++ 14.0 or greater**.
+> You can download the **Microsoft C++ Build Tools** [here](https://visualstudio.microsoft.com/visual-cpp-build-tools/).
+
 📝 By default, `talk2knowledgegraphs` includes a small subset of the PrimeKG knowledge graph, allowing users to start interacting with it out of the box.  
 To switch to a different knowledge graph or use your own, refer to the [deployment guide](https://virtualpatientengine.github.io/AIAgents4Pharma/talk2knowledgegraphs/deployment/).
 
-For detailed instructions on each agent, please refer to their respective modules.
+**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
+
+_Please note that this will create a new tracing project in your Langsmith
+account with the name `T2X-xxxx`, where `X` can be `AA4P` (Main Agent),
+`B` (Biomodels), `S` (Scholars), `KG` (KnowledgeGraphs), or `C` (Cells).
+If you skip the previous step, it will default to the name `default`.
+`xxxx` will be the 4-digit ID created for the session._
 
 #### Option 3: pip (beta-release)
 

{aiagents4pharma-1.40.0 → aiagents4pharma-1.41.0}/README.md

@@ -43,103 +43,12 @@ Our toolkit currently consists of the following agents:
 
 _We now have all the agents available on Docker Hub._
 
-##### **To run Talk2AIAgents4Pharma / Talk2KnowledgeGraphs**
-
-If your machine has NVIDIA GPU(s), please install the following this:
-- [nvidia-cuda-toolkit](https://developer.nvidia.com/cuda-toolkit)
-- [nvidia-container-toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/1.17.8/install-guide.html) (required for GPU support with Docker; enables containers to access NVIDIA GPUs for accelerated computing). After installing `nvidia-container-toolkit`, please restart Docker to ensure GPU support is enabled.
-
-##### **1. Download docker-compose.yml, .env.example and startup.sh from GitHub**
-
-###### Talk2Agents4Pharma
-
-```sh
-mkdir talk2aiagents4pharma && cd talk2aiagents4pharma && wget https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/docker-compose.yml https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/.env.example https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2aiagents4pharma/startup.sh
-```
-
-###### Talk2KnowledgeGraphs
-
-```sh
-mkdir talk2knowledgegraphs && cd talk2knowledgegraphs && wget https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/docker-compose.yml https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/.env.example https://raw.githubusercontent.com/VirtualPatientEngine/AIAgents4Pharma/main/aiagents4pharma/talk2knowledgegraphs/startup.sh
-```
-
-##### **2. Setup environment variables**
-
-Copy and configure your `.env` file:
-
-```sh
-cp .env.example .env
-```
-
-Then edit `.env` and add your API keys:
-
-```env
-OPENAI_API_KEY=...                  # Required for both agents
-NVIDIA_API_KEY=...                  # Required for both agents
-LANGCHAIN_TRACING_V2=true           # Optional for both agents
-LANGCHAIN_API_KEY=...               # Optional for both agents
-```
-
-[Additional Notes for Windows Users](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#notes-for-windows-users)
-
-##### **3. Start the application**
-
-```sh
-chmod +x startup.sh
-./startup.sh        # Add --cpu flag to force CPU mode if needed
-```
-
-[More about startup script](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#about-startupsh)
-
-##### **To Run Talk2Biomodels / Talk2Scholars**
-
-###### Talk2Biomodels
-
-```docker
-docker run -d \
---name talk2biomodels \
--e OPENAI_API_KEY=<your_openai_api_key> \
--e NVIDIA_API_KEY=<your_nvidia_api_key> \
--p 8501:8501 \
-virtualpatientengine/talk2biomodels
-```
-
-###### Talk2Scholars
-
-```docker
-docker run -d \
---name talk2scholars \
--e OPENAI_API_KEY=<your_openai_api_key> \
--e ZOTERO_API_KEY=<your_zotero_api_key> \
--e ZOTERO_USER_ID=<your_zotero_user_id> \
--e NVIDIA_API_KEY=<your_nvidia_api_key> \
--p 8501:8501 \
-virtualpatientengine/talk2scholars
-```
-
-##### **4. Access the Web UI**
-
-Once started, the agent is available at:
-
-```
-http://localhost:8501
-```
-
-To use **Talk2AIAgents4Pharma** or **Talk2KnowledgeGraphs**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-To use **Talk2BioModels** or **Talk2Scholars**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-Only for **Talk2Scholars**, you also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(For all other agents, the Zotero key is not required.)_
-
-If you are using docker on Windows, please follow these [Windows Setup Notes](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#notes-for-windows-users).
-
-**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
-
-[More on running multiple agents simultaneously](https://github.com/VirtualPatientEngine/AIAgents4Pharma/blob/main/aiagents4pharma/talk2aiagents4pharma/install.md#to-run-multiple-agents-simultaneously)
-
-📝 By default, `talk2knowledgegraphs` includes a small subset of the PrimeKG knowledge graph, allowing users to start interacting with it out of the box.  
-To switch to a different knowledge graph or use your own, refer to the [deployment guide](https://virtualpatientengine.github.io/AIAgents4Pharma/talk2knowledgegraphs/deployment/).
+Choose your agent below for detailed Docker instructions:
 
+- [Talk2AIAgents4Pharma](aiagents4pharma/talk2aiagents4pharma/install.md)
+- [Talk2KnowledgeGraphs](aiagents4pharma/talk2knowledgegraphs/install.md)
+- [Talk2BioModels](aiagents4pharma/talk2biomodels/install.md)
+- [Talk2Scholars](aiagents4pharma/talk2scholars/install.md)
 
 #### Option 2: git (for developers and contributors)
 
@@ -170,23 +79,6 @@ export LANGCHAIN_TRACING_V2=true    # Optional for all agents
 export LANGCHAIN_API_KEY=...        # Optional for all agents
 ```
 
-To use **Talk2AIAgents4Pharma**, **Talk2BioModels**, **Talk2KnowledgeGraphs**, or **Talk2Scholars**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
-
-Only for **Talk2Scholars**, you also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(For all other agents, the Zotero key is not required.)_
-
-To use **Talk2Scholars**, you must have **FAISS** installed through **Conda**. Follow installation instructions for your OS [here](https://github.com/VirtualPatientEngine/AIAgents4Pharma/tree/main/aiagents4pharma/talk2scholars/install.md).
-
-Additionally on **Windows**, the `pcst_fast 1.0.10` library requires **Microsoft Visual C++ 14.0 or greater**.  
-You can download the **Microsoft C++ Build Tools** [here](https://visualstudio.microsoft.com/visual-cpp-build-tools/).
-
-**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
-
-_Please note that this will create a new tracing project in your Langsmith
-account with the name `T2X-xxxx`, where `X` can be `AA4P` (Main Agent),
-`B` (Biomodels), `S` (Scholars), `KG` (KnowledgeGraphs), or `C` (Cells).
-If you skip the previous step, it will default to the name `default`.
-`xxxx` will be the 4-digit ID created for the session._
-
 4. **Launch the app:**
 
 ```sh
@@ -201,10 +93,28 @@ _Replace `<agent>` with the agent name you are interested to launch:_
 - `talk2scholars`
 - `talk2cells`
 
+If your machine has NVIDIA GPU(s), please install the following this:
+
+- [nvidia-cuda-toolkit](https://developer.nvidia.com/cuda-toolkit)
+- [nvidia-container-toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/1.17.8/install-guide.html) (required for GPU support with Docker; enables containers to access NVIDIA GPUs for accelerated computing). After installing `nvidia-container-toolkit`, please restart Docker to ensure GPU support is enabled.
+
+To use the **Agents**, you need a free **NVIDIA API key**. Create an account and apply for free credits [here](https://build.nvidia.com/explore/discover).
+
+**Talk2Scholars** requires Milvus to be set up as the vector database — install Milvus depending on your setup by following the official instructions for [CPU](https://milvus.io/docs/install_standalone-docker-compose.md) or [GPU](https://milvus.io/docs/install_standalone-docker-compose-gpu.md). You will also need a **Zotero API key**, which you can generate [here](https://www.zotero.org/user/login#applications). _(The Zotero key is only required for Talk2Scholars; all other agents do not need it.)_
+
+> Additionally on **Windows**, the `pcst_fast 1.0.10` library requires **Microsoft Visual C++ 14.0 or greater**.
+> You can download the **Microsoft C++ Build Tools** [here](https://visualstudio.microsoft.com/visual-cpp-build-tools/).
+
 📝 By default, `talk2knowledgegraphs` includes a small subset of the PrimeKG knowledge graph, allowing users to start interacting with it out of the box.  
 To switch to a different knowledge graph or use your own, refer to the [deployment guide](https://virtualpatientengine.github.io/AIAgents4Pharma/talk2knowledgegraphs/deployment/).
 
-For detailed instructions on each agent, please refer to their respective modules.
+**LangSmith** support is optional. To enable it, create an API key [here](https://docs.smith.langchain.com/administration/how_to_guides/organization_management/create_account_api_key).
+
+_Please note that this will create a new tracing project in your Langsmith
+account with the name `T2X-xxxx`, where `X` can be `AA4P` (Main Agent),
+`B` (Biomodels), `S` (Scholars), `KG` (KnowledgeGraphs), or `C` (Cells).
+If you skip the previous step, it will default to the name `default`.
+`xxxx` will be the 4-digit ID created for the session._
 
 #### Option 3: pip (beta-release)
 

aiagents4pharma-1.41.0/aiagents4pharma/talk2scholars/configs/agents/talk2scholars/s2_agent/default.yaml

@@ -0,0 +1,9 @@
+_target_: agents.s2_agent.get_app
+s2_agent: |
+  You are the S2 Agent.
+
+  You are responsible for searching academic papers, getting recommendations based on the searched articles, and displaying the results.
+
+  IMPORTANT INSTRUCTION FOR AGENT BEHAVIOR:
+  If the user's request involves extracting paper IDs to download papers, your task is only to extract those IDs using the `query_dataframe`. Do not attempt to download the paper yourself or call any other tools after extracting the IDs.
+  Once the IDs are successfully extracted, immediately pause execution and return control to the main agent. The main agent is responsible for invoking the appropriate tool or sub-agent to handle the paper download.

aiagents4pharma-1.41.0/aiagents4pharma/talk2scholars/configs/tools/question_and_answer/default.yaml

@@ -0,0 +1,62 @@
+# Configuration for the PDF question_and_answer Tool - Traditional RAG Pipeline with GPU Support
+
+# Milvus vector database settings
+milvus:
+  # Connection settings
+  host: ${oc.env:MILVUS_HOST,localhost} # Changed default from 127.0.0.1 to localhost
+  port: ${oc.env:MILVUS_PORT,19530}
+
+  # Database and collection settings
+  db_name: ${oc.env:MILVUS_DB_NAME,pdf_rag_db}
+  collection_name: ${oc.env:MILVUS_COLLECTION_NAME,pdf_rag_documents}
+
+  # Ensure collection persists across restarts
+  consistency_level: "Strong"
+  embedding_dim: 768
+
+# Document processing settings
+chunk_size: 1200 # Number of characters per text chunk
+chunk_overlap: 200 # Overlap between adjacent chunks
+
+# Parallel processing settings
+embedding_batch_size: 1500 # Number of chunks to embed in a single API call
+max_parallel_pdfs: 10 # Maximum number of PDFs to process in parallel
+
+# Traditional RAG Pipeline Settings
+# Step 1: Initial retrieval (cast wide net)
+initial_retrieval_k: 100 # Number of chunks to retrieve before reranking
+mmr_diversity: 0.8 # MMR diversity parameter (0=max diversity, 1=max relevance)
+
+# Step 2: Reranking settings
+top_k_chunks: 25 # Final number of chunks after reranking
+reranker:
+  model: "nvidia/nv-rerankqa-mistral-4b-v3"
+  api_key: ${oc.env:NVIDIA_API_KEY}
+
+# Answer generation settings
+prompt_template: |
+  You are a scientific research assistant specialized in reading and extracting information from research papers.
+  Your role is to answer questions by retrieving relevant information from the provided context.
+
+  - Provide detailed, structured, and well-argued explanations—not just brief summaries.
+  - Cite specific sources using only the title of the paper.
+  - If the context is insufficient, clearly state that more information is needed.
+
+  Context:
+  {context}
+
+  Question: {question}
+
+  Your answer should be comprehensive, accurate, and clearly structured for a scientific audience.
+
+# GPU Detection and Performance Settings
+gpu_detection:
+  # Timeout for GPU detection command (seconds)
+  detection_timeout: 10
+
+  # Log GPU detection results
+  log_detection: true
+
+  # Force CPU mode even if GPU is detected (for testing CPU Milvus)
+  # Uncomment the line below to force CPU mode:
+  # force_cpu_mode: true

aiagents4pharma-1.41.0/aiagents4pharma/talk2scholars/tests/test_nvidia_nim_reranker.py

@@ -0,0 +1,127 @@
+"""
+Unit tests for NVIDIA NIM reranker error handling in nvidia_nim_reranker.py
+"""
+
+from unittest.mock import MagicMock, patch
+
+import pytest
+from langchain_core.documents import Document
+
+from aiagents4pharma.talk2scholars.tools.pdf.utils import nvidia_nim_reranker
+from aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker import (
+    rerank_chunks,
+)
+
+
+@pytest.fixture(name="chunks_fixture")
+def fixture_chunks():
+    """chunks_fixture fixture to simulate PDF chunks."""
+    return [
+        Document(
+            page_content=f"chunk {i}",
+            metadata={"paper_id": f"P{i%2}", "relevance_score": 0.9 - 0.01 * i},
+        )
+        for i in range(10)
+    ]
+
+
+def test_rerank_chunks_short_input(chunks_fixture):
+    """rerank_chunks with fewer chunks than top_k should return original."""
+    result = rerank_chunks(
+        chunks_fixture[:3], "What is cancer?", config=MagicMock(), top_k=5
+    )
+    assert result == chunks_fixture[:3]
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker.logger")
+def test_rerank_chunks_missing_api_key_logs_and_raises(mock_logger, chunks_fixture):
+    """
+    If config.reranker.api_key is None:
+      - logger.error(...) should be called
+      - rerank_chunks should raise ValueError
+    """
+    mock_config = MagicMock()
+    mock_config.reranker.api_key = None
+
+    with pytest.raises(
+        ValueError,
+        match="Configuration 'reranker.api_key' must be set for reranking",
+    ):
+        rerank_chunks(chunks_fixture, "What is cancer?", config=mock_config, top_k=5)
+
+    mock_logger.error.assert_called_once_with(
+        "No NVIDIA API key found in configuration for reranking"
+    )
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker.NVIDIARerank")
+def test_rerank_chunks_success(mock_reranker_cls, chunks_fixture):
+    """rerank_chunks with successful reranking."""
+    reranker_instance = MagicMock()
+    reranker_instance.compress_documents.return_value = list(reversed(chunks_fixture))
+    mock_reranker_cls.return_value = reranker_instance
+
+    mock_config = MagicMock()
+    mock_config.reranker.api_key = "test_key"
+    mock_config.reranker.model = "test_model"
+
+    result = rerank_chunks(
+        chunks_fixture, "Explain mitochondria.", config=mock_config, top_k=5
+    )
+
+    assert isinstance(result, list)
+    assert result == list(reversed(chunks_fixture))[:5]
+    reranker_instance.compress_documents.assert_called_once_with(
+        query="Explain mitochondria.", documents=chunks_fixture
+    )
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker.NVIDIARerank")
+def test_rerank_chunks_reranker_fails_raises_and_calls_compress(
+    mock_reranker_cls, chunks_fixture
+):
+    """
+    If NVIDIARerank.compress_documents raises RuntimeError:
+      - rerank_chunks should propagate the RuntimeError
+      - and compress_documents should have been called
+    """
+    reranker_instance = MagicMock()
+    reranker_instance.compress_documents.side_effect = RuntimeError("API failure")
+    mock_reranker_cls.return_value = reranker_instance
+
+    mock_config = MagicMock()
+    mock_config.reranker.api_key = "valid_key"
+    mock_config.reranker.model = "reranker"
+
+    with pytest.raises(RuntimeError, match="API failure"):
+        rerank_chunks(
+            chunks_fixture, "How does light affect plants?", config=mock_config, top_k=3
+        )
+
+    reranker_instance.compress_documents.assert_called_once_with(
+        query="How does light affect plants?", documents=chunks_fixture
+    )
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker.logger")
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.nvidia_nim_reranker.NVIDIARerank")
+def test_rerank_chunks_debug_block_triggered(
+    mock_reranker_cls, mock_logger, chunks_fixture
+):
+    """rerank_chunks should log debug info if debug logging is enabled."""
+    mock_logger.isEnabledFor.return_value = True
+
+    reranker_instance = MagicMock()
+    reranker_instance.compress_documents.return_value = chunks_fixture
+    mock_reranker_cls.return_value = reranker_instance
+
+    mock_config = MagicMock()
+    mock_config.reranker.api_key = "abc"
+    mock_config.reranker.model = "mymodel"
+
+    result = nvidia_nim_reranker.rerank_chunks(
+        chunks_fixture * 2, "Test query", mock_config, top_k=3
+    )
+
+    assert result == chunks_fixture[:3]
+    assert mock_logger.debug.called

aiagents4pharma-1.41.0/aiagents4pharma/talk2scholars/tests/test_pdf_answer_formatter.py

@@ -0,0 +1,66 @@
+"""answer_formatter tests."""
+
+from unittest.mock import patch
+import pytest
+
+from aiagents4pharma.talk2scholars.tools.pdf.utils.answer_formatter import format_answer
+
+
+@pytest.fixture(name="base_args")
+def _base_args():
+    """base_args fixture to provide common arguments for tests."""
+    return {
+        "question": "What is the conclusion?",
+        "chunks": [{"content": "chunk1"}, {"content": "chunk2"}],
+        "llm": "mock_llm",
+        "articles": {
+            "paper1": {"Title": "Paper One"},
+            "paper2": {"Title": "Paper Two"},
+        },
+        "config": {"key": "value"},
+        "call_id": "test_call_123",
+        "has_gpu": True,
+    }
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.answer_formatter.generate_answer")
+def test_format_answer_with_sources(mock_generate_answer, base_args):
+    """test format_answer with sources."""
+    mock_generate_answer.return_value = {
+        "output_text": "This is the generated answer.",
+        "papers_used": ["paper1", "paper2"],
+    }
+
+    result = format_answer(**base_args)
+
+    assert "This is the generated answer." in result
+    assert "Sources:" in result
+    assert "- Paper One" in result
+    assert "- Paper Two" in result
+    mock_generate_answer.assert_called_once()
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.answer_formatter.generate_answer")
+def test_format_answer_no_sources(mock_generate_answer, base_args):
+    """test format_answer with no sources."""
+    mock_generate_answer.return_value = {
+        "output_text": "No sources were used.",
+        "papers_used": [],  # No papers used
+    }
+
+    result = format_answer(**base_args)
+
+    assert result == "No sources were used."  # No sources section expected
+    mock_generate_answer.assert_called_once()
+
+
+@patch("aiagents4pharma.talk2scholars.tools.pdf.utils.answer_formatter.generate_answer")
+def test_format_answer_missing_output_text(mock_generate_answer, base_args):
+    """test format_answer with missing output text."""
+    mock_generate_answer.return_value = {"papers_used": ["paper1"]}
+
+    result = format_answer(**base_args)
+
+    assert result.startswith("No answer generated.")
+    assert "Sources:" in result
+    mock_generate_answer.assert_called_once()