DEFAULT_DOCUMENT_PROMPT

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-    "core": "0.0.21"
+    "core": "0.0.22"
 }

README.md

@@ -20,7 +20,7 @@ Quivr, helps you build your second brain, utilizes the power of GenerativeAI to
 
 >We take care of the RAG so you can focus on your product. Simply install quivr-core and add it to your project. You can now ingest your files and ask questions.*
 
-**We will be improving the RAG and adding more features everything, stay tuned!**
+**We will be improving the RAG and adding more features, stay tuned!**
 
 
 This is the core of Quivr, the brain of Quivr.com.
@@ -53,7 +53,7 @@ Ensure you have the following installed:
 
 - **Step 2**: Create a RAG with 5 lines of code
 
-```python
+  ```python
   import tempfile
 
   from quivr_core import Brain
@@ -72,7 +72,7 @@ Ensure you have the following installed:
               "what is gold? asnwer in french"
           )
           print("answer:", answer)
-```
+  ```
 
 ## Examples 
 

core/CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [0.0.22](https://github.com/QuivrHQ/quivr/compare/core-0.0.21...core-0.0.22) (2024-10-21)
+
+
+### Features
+
+* **ask:** non-streaming now calls streaming ([#3409](https://github.com/QuivrHQ/quivr/issues/3409)) ([e71e46b](https://github.com/QuivrHQ/quivr/commit/e71e46bcdfbab0d583aef015604278343fd46c6f))
+
 ## [0.0.21](https://github.com/QuivrHQ/quivr/compare/core-0.0.20...core-0.0.21) (2024-10-21)
 
 

core/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "quivr-core"
-version = "0.0.21"
+version = "0.0.22"
 description = "Quivr core RAG package"
 authors = [
     { name = "Stan Girard", email = "[email protected]" }

core/quivr_core/brain/brain.py

@@ -545,34 +545,28 @@ def ask(
         print(answer.answer)
         ```
         """
-        llm = self.llm
-
-        # If you passed a different llm model we'll override the brain  one
-        if retrieval_config:
-            if retrieval_config.llm_config != self.llm.get_config():
-                llm = LLMEndpoint.from_config(config=retrieval_config.llm_config)
-        else:
-            retrieval_config = RetrievalConfig(llm_config=self.llm.get_config())
-
-        if rag_pipeline is None:
-            rag_pipeline = QuivrQARAGLangGraph
-
-        rag_instance = rag_pipeline(
-            retrieval_config=retrieval_config, llm=llm, vector_store=self.vector_db
-        )
+        async def collect_streamed_response():
+            full_answer = ""
+            async for response in self.ask_streaming(
+                question=question,
+                retrieval_config=retrieval_config,
+                rag_pipeline=rag_pipeline,
+                list_files=list_files,
+                chat_history=chat_history
+            ):
+                full_answer += response.answer
+            return full_answer
+
+        # Run the async function in the event loop
+        loop = asyncio.get_event_loop()
+        full_answer = loop.run_until_complete(collect_streamed_response())
 
         chat_history = self.default_chat if chat_history is None else chat_history
-        list_files = [] if list_files is None else list_files
-
-        parsed_response = rag_instance.answer(
-            question=question, history=chat_history, list_files=list_files
-        )
-
         chat_history.append(HumanMessage(content=question))
-        chat_history.append(AIMessage(content=parsed_response.answer))
+        chat_history.append(AIMessage(content=full_answer))
 
-        # Save answer to the chat history
-        return parsed_response
+        # Return the final response
+        return ParsedRAGResponse(answer=full_answer)
 
     async def ask_streaming(
         self,
@@ -635,3 +629,4 @@ async def ask_streaming(
         chat_history.append(HumanMessage(content=question))
         chat_history.append(AIMessage(content=full_answer))
         yield response
+

core/quivr_core/config.py

@@ -233,7 +233,7 @@ class LLMEndpointConfig(QuivrBaseConfig):
 
     Attributes:
         supplier (DefaultModelSuppliers): The LLM provider (default: OPENAI).
-        model (str): The specific model to use (default: "gpt-3.5-turbo-0125").
+        model (str): The specific model to use (default: "gpt-4o").
         context_length (int | None): The maximum context length for the model.
         tokenizer_hub (str | None): The tokenizer to use for this model.
         llm_base_url (str | None): Base URL for the LLM API.
@@ -247,7 +247,7 @@ class LLMEndpointConfig(QuivrBaseConfig):
     """
 
     supplier: DefaultModelSuppliers = DefaultModelSuppliers.OPENAI
-    model: str = "gpt-3.5-turbo-0125"
+    model: str = "gpt-4o"
     context_length: int | None = None
     tokenizer_hub: str | None = None
     llm_base_url: str | None = None

core/quivr_core/utils.py

@@ -26,7 +26,7 @@ def model_supports_function_calling(model_name: str):
         "gpt-4",
         "gpt-4-1106-preview",
         "gpt-4-0613",
-        "gpt-3.5-turbo-0125",
+        "gpt-4o",
         "gpt-3.5-turbo-1106",
         "gpt-3.5-turbo-0613",
         "gpt-4-0125-preview",

core/tests/rag_config.yaml

@@ -27,7 +27,7 @@ retrieval_config:
     supplier: "openai"
 
     # The model to use for the LLM for the given supplier
-    model: "gpt-3.5-turbo-0125"
+    model: "gpt-4o"
 
     max_input_tokens: 2000
 

core/tests/rag_config_workflow.yaml

@@ -40,7 +40,7 @@ retrieval_config:
     supplier: "openai"
 
     # The model to use for the LLM for the given supplier
-    model: "gpt-3.5-turbo-0125"
+    model: "gpt-4o"
 
     max_input_tokens: 2000
 

core/tests/test_config.py

@@ -5,7 +5,7 @@ def test_default_llm_config():
     config = LLMEndpointConfig()
 
     assert config.model_dump(exclude={"llm_api_key"}) == LLMEndpointConfig(
-        model="gpt-3.5-turbo-0125",
+        model="gpt-4o",
         llm_base_url=None,
         llm_api_key=None,
         max_input_tokens=2000,

docs/docs/index.md

@@ -1,41 +1,68 @@
 # Welcome to Quivr Documentation
 
-Welcome to the documentation of Quivr! This is the place where you'll find help, guidance and support for collaborative software development. Whether you're involved in an open-source community or a large software team, these resources should get you up and running quickly!
+Quivr, helps you build your second brain, utilizes the power of GenerativeAI to be your personal assistant !
 
-[Quivr](https://quivr.app) is your **Second Brain** that can act as your **personal assistant**. Quivr is a platform that enables the creation of AI assistants, referred to as "Brain". These assistants are designed with specialized capabilities. Some can connect to specific data sources, allowing users to interact directly with the data. Others serve as specialized tools for particular use cases, powered by Rag technology. These tools process specific inputs to generate practical outputs, such as summaries, translations, and more.
+## Key Features 🎯
 
-## Quick Links
+- **Opiniated RAG**: We created a RAG that is opinionated, fast and efficient so you can focus on your product
+- **LLMs**: Quivr works with any LLM, you can use it with OpenAI, Anthropic, Mistral, Gemma, etc.
+- **Any File**: Quivr works with any file, you can use it with PDF, TXT, Markdown, etc and even add your own parsers.
+- **Customize your RAG**: Quivr allows you to customize your RAG, add internet search, add tools, etc.
+- **Integrations with Megaparse**: Quivr works with [Megaparse](https://github.com/quivrhq/megaparse), so you can ingest your files with Megaparse and use the RAG with Quivr.
 
-- [Video Installation](https://dub.sh/quivr-demo)
+>We take care of the RAG so you can focus on your product. Simply install quivr-core and add it to your project. You can now ingest your files and ask questions.*
 
-!!! note
-    **Our goal** is to make Quivr the **best personal assistant** that is powered by your knowledge and your applications 🔥 
+**We will be improving the RAG and adding more features everything, stay tuned!**
 
-## What does it do?
 
-<div style="text-align: center;">
-  <video width="640" height="480" controls>
-    <source src="https://quivr-cms.s3.eu-west-3.amazonaws.com/singlestore_demo_quivr_232893659c.mp4" type="video/mp4">
-    Your browser does not support the video tag.
-  </video>
-</div>
+This is the core of Quivr, the brain of Quivr.com.
 
-## How to get started? 👀
+<!-- ## Demo Highlight 🎥
 
-!!! tip
-    It takes less than **5 seconds** to get started with Quivr. You can even use your Google account to sign up.
+https://github.com/quivrhq/quivr/assets/19614572/a6463b73-76c7-4bc0-978d-70562dca71f5 -->
 
-1. **Create an account**: Go to [Quivr](https://quivr.app) and create an account.
-2. **Create a Brain**: Let us guide you to create your first brain!
-3. **Feed this Brain**: Add documentation and/or URLs to feed your brain.
-4. **Ask Questions to your Brain**: Ask your Brain questions about the knowledge that you provide.
+## Getting Started 🚀
 
-## Empowering Innovation with Foundation Models & Generative AI
+You can find everything on the [documentation](https://core.quivr.app/).
 
-As a Leader in AI, Quivr leverages Foundation Models and Generative AI to empower businesses to achieve gains through Innovation.
+### Prerequisites 📋
 
-- 50k+ users
-- 6k+ companies
-- 35k+ github stars
-- Top 100 open-source
+Ensure you have the following installed:
+
+- Python 3.10 or newer
+
+### 30 seconds Installation 💽
+
+
+- **Step 1**: Install the package
+
+
+
+  ```bash
+  pip install quivr-core # Check that the installation worked
+  ```
+
+
+- **Step 2**: Create a RAG with 5 lines of code
+
+  ```python
+    import tempfile
+
+    from quivr_core import Brain
+
+    if __name__ == "__main__":
+        with tempfile.NamedTemporaryFile(mode="w", suffix=".txt") as temp_file:
+            temp_file.write("Gold is a liquid of blue-like colour.")
+            temp_file.flush()
+
+            brain = Brain.from_files(
+                name="test_brain",
+                file_paths=[temp_file.name],
+            )
+
+            answer = brain.ask(
+                "what is gold? asnwer in french"
+            )
+            print("answer:", answer)
+  ```
 

docs/docs/quickstart.md

@@ -70,13 +70,13 @@ brain = Brain.from_files(name = "my smart brain",
 
 Note : [Embeddings](https://python.langchain.com/docs/integrations/text_embedding/) is a langchain class that lets you chose from a large variety of embedding models. Please check out the following docs to know the panel of models you can try.
 
-## Launch with Streamlit
+## Launch with Chainlit
 
 If you want to quickly launch an interface with streamlit, you can simply do at the root of the project :
 ```bash
 cd examples/chatbot /
 rye sync /
-rye run streamlit run examples/chatbot/main.py /
+rye run chainlit run chainlit.py
 ```
 For more detail, go in [examples/chatbot/chainlit.md](https://github.com/QuivrHQ/quivr/tree/main/examples/chatbot)
 

docs/mkdocs.yml

@@ -45,6 +45,7 @@ theme:
         name: Switch to system preference
 
 plugins:
+  - search
   - mkdocstrings:
       default_handler: python
       handlers:
@@ -58,7 +59,6 @@ plugins:
 nav:
   - Home:
       - index.md
-      - installation.md
       - quickstart.md
       - Brain:
           - brain/index.md