pygpt-net 2.4.45__py3-none-any.whl → 2.4.46__py3-none-any.whl

CHANGELOG.md

@@ -1,5 +1,11 @@
 # CHANGELOG
 
+## 2.4.46 (2024-12-16)
+
+- Added a new tab in Settings: "API Keys", where the API keys configuration for Google and Anthropic models has been relocated.
+- Introduced a new mode in "Chat with Files": "Retrieve Only", which allows for retrieving raw documents from the index.
+- Fixed a bug related to tool calls in the Gemini provider when using Chat with Files mode.
+
 ## 2.4.45 (2024-12-16)
 
 - Enhanced web data loaders UI.

README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.4.45** | build: **2024.12.16** | Python: **>=3.10, <3.12**
+Release: **2.4.46** | build: **2024.12.16** | Python: **>=3.10, <3.12**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -329,12 +329,12 @@ During the initial launch, you must configure your API key within the applicatio
 To do so, navigate to the menu:
 
 ``` ini
-Config -> Settings...
+Config -> Settings -> API Keys
 ```
 
 and then paste the API key into the `OpenAI API KEY` field.
 
-![v2_settings](https://github.com/user-attachments/assets/003b0f86-8225-4478-8525-fb9324ac5c88)
+![v2_api_keys](https://github.com/user-attachments/assets/8564add8-364b-471d-80d5-7e99ae77e129)
 
 The API key can be obtained by registering on the OpenAI website:
 
@@ -354,6 +354,8 @@ Your API keys will be available here:
 
 This mode in **PyGPT** mirrors `ChatGPT`, allowing you to chat with models such as `o1`, `GPT-4`, `GPT-4o` and `GPT-3.5`. It works by using the `ChatCompletion` OpenAI API.
 
+**Tip: This mode directly uses the OpenAI API. If you want to use models other than GPT (such as Gemini, Claude, or Llama3), use `Chat with Files` mode.**
+
 The main part of the interface is a chat window where you see your conversations. Below it is a message box for typing. On the right side, you can set up or change the model and system prompt. You can also save these settings as presets to easily switch between models or tasks.
 
 Above where you type your messages, the interface shows you the number of tokens your message will use up as you type it – this helps to keep track of usage. There is also a feature to attach and upload files in this area. Go to the `Files and Attachments` section for more information on how to use attachments.
@@ -2723,7 +2725,17 @@ Config -> Settings...
 
 **General**
 
-- `OpenAI API KEY`: The personal API key you'll need to enter into the application for it to function.
+- `Minimize to tray on exit`: Minimize to tray icon on exit. Tray icon enabled is required for this option to work. Default: False.
+
+- `Render engine`: chat output render engine: `WebEngine / Chromium` - for full HTML/CSS and `Legacy (markdown)` for legacy, simple markdown CSS output. Default: WebEngine / Chromium.
+
+- `OpenGL hardware acceleration`: enables hardware acceleration in `WebEngine / Chromium` renderer.  Default: False.
+
+- `Application environment (os.environ)`: Additional environment vars to set on application start.
+
+**API Keys**
+
+- `OpenAI API KEY`: Required for the OpenAI API. If you wish to use custom endpoints or local APIs, then you may enter any value here.
 
 - `OpenAI ORGANIZATION KEY`: The organization's API key, which is optional for use within the application.
 
@@ -2731,13 +2743,15 @@ Config -> Settings...
 
 - `Proxy address`: Proxy address to be used for connection; supports HTTP/SOCKS.
 
-- `Minimize to tray on exit`: Minimize to tray icon on exit. Tray icon enabled is required for this option to work. Default: False.
+- `Google API KEY`: Required for the Google API and Gemini models.
 
-- `Render engine`: chat output render engine: `WebEngine / Chromium` - for full HTML/CSS and `Legacy (markdown)` for legacy, simple markdown CSS output. Default: WebEngine / Chromium.
+- `Anthropic API KEY`: Required for the Anthropic API and Claude models.
 
-- `OpenGL hardware acceleration`: enables hardware acceleration in `WebEngine / Chromium` renderer.  Default: False.
+- `HuggingFace API KEY`: Required for the HuggingFace API.
 
-- `Application environment (os.environ)`: Additional environment vars to set on application start.
+- `OpenAI API version`: Azure OpenAI API version, e.g. 2023-07-01-preview
+
+- `Azure OpenAI API endpoint`: Azure OpenAI API endpoint, https://<your-resource-name>.openai.azure.com/
 
 **Layout**
 
@@ -3938,6 +3952,12 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.4.46 (2024-12-16)**
+
+- Added a new tab in Settings: "API Keys", where the API keys configuration for Google and Anthropic models has been relocated.
+- Introduced a new mode in "Chat with Files": "Retrieve Only", which allows for retrieving raw documents from the index.
+- Fixed a bug related to tool calls in the Gemini provider when using Chat with Files mode.
+
 **2.4.45 (2024-12-16)**
 
 - Enhanced web data loaders UI.

pygpt_net/CHANGELOG.txt

@@ -1,3 +1,9 @@
+2.4.46 (2024-12-16)
+
+- Added a new tab in Settings: "API Keys", where the API keys configuration for Google and Anthropic models has been relocated.
+- Introduced a new mode in "Chat with Files": "Retrieve Only", which allows for retrieving raw documents from the index.
+- Fixed a bug related to tool calls in the Gemini provider when using Chat with Files mode.
+
 2.4.45 (2024-12-16)
 
 - Enhanced web data loaders UI.

pygpt_net/__init__.py

@@ -6,14 +6,14 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.16 02:00:00                  #
+# Updated Date: 2024.12.16 20:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2024, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.4.45"
+__version__ = "2.4.46"
 __build__ = "2024.12.16"
 __maintainer__ = "Marcin Szczygliński"
 __github__ = "https://github.com/szczyglis-dev/py-gpt"

pygpt_net/controller/chat/text.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 00:00:00                  #
+# Updated Date: 2024.12.16 20:00:00                  #
 # ================================================== #
 
 from typing import Optional
@@ -83,10 +83,13 @@ class Text:
         base_mode = mode  # store parent mode
         functions = []  # functions to call
         tools_outputs = []  # tools outputs (assistant only)
+        idx_mode = self.window.core.config.get('llama.idx.mode')
 
         # o1 models: disable stream mode
         if model.startswith("o1") or mode in [MODE_AGENT_LLAMA, MODE_AUDIO]:
             stream_mode = False
+        if mode in [MODE_LLAMA_INDEX] and idx_mode == "retrieval":
+            stream_mode = False
 
         # create ctx item
         ctx = CtxItem()
@@ -234,7 +237,7 @@ class Text:
                 file_ids=self.window.controller.files.uploaded_ids,  # uploaded files IDs
                 assistant_id=assistant_id,
                 idx=self.window.controller.idx.current_idx,  # current idx
-                idx_mode=self.window.core.config.get('llama.idx.mode'),  # llama index mode (chat or query)
+                idx_mode=idx_mode,  # llama index mode (chat or query)
                 external_functions=functions,  # external functions
                 tools_outputs=tools_outputs,  # if not empty then will submit outputs to assistant
                 max_tokens=max_tokens,  # max output tokens

pygpt_net/controller/idx/__init__.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 08:00:00                  #
+# Updated Date: 2024.12.16 20:00:00                  #
 # ================================================== #
 
 import datetime
@@ -65,6 +65,7 @@ class Idx:
         return [
             {"chat": trans('toolbox.llama_index.mode.chat')},
             {"query": trans('toolbox.llama_index.mode.query')},
+            {"retrieval": trans('toolbox.llama_index.mode.retrieval')},
         ]
 
     def select_mode(self, mode: str):

pygpt_net/controller/settings/__init__.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 00:00:00                  #
+# Updated Date: 2024.12.16 20:00:00                  #
 # ================================================== #
 
 import os
@@ -230,7 +230,7 @@ class Settings:
 
     def welcome_settings(self):
         """Open settings at first launch (if no API key yet)"""
-        self.open_section("general")
+        self.open_section("api_keys")
         self.window.ui.config['config']['api_key'].setFocus()
         self.window.ui.dialogs.close('info.start')
 

pygpt_net/core/idx/chat.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2024.12.14 22:00:00                  #
+# Updated Date: 2024.12.16 20:00:00                  #
 # ================================================== #
 
 import json
@@ -61,6 +61,11 @@ class Chat:
                 context=context,
                 extra=extra,
             )
+        elif idx_mode == "retrieval":  # retrieval mode
+            return self.retrieval(
+                context=context,
+                extra=extra,
+            )
 
         # if not raw, check if chat mode is available
         if MODE_CHAT in model.llama_index['mode']:
@@ -163,6 +168,50 @@ class Chat:
             return True
         return False
 
+    def retrieval(
+            self,
+            context: BridgeContext,
+            extra: Optional[Dict[str, Any]] = None
+    ) -> bool:
+        """
+        Retrieve documents from index only
+
+        :param context: Bridge context
+        :param extra: Extra arguments
+        :return: True if success
+        """
+        idx = context.idx
+        model = context.model
+        ctx = context.ctx
+        query = ctx.input  # user input
+        verbose = self.window.core.config.get("log.llama", False)
+
+        self.log("Retrieval...")
+        self.log("Idx: {}, retrieve only: {}".format(
+            idx,
+            query,
+        ))
+
+        index, service_context = self.get_index(idx, model)
+        retriever = index.as_retriever()
+        nodes = retriever.retrieve(query)
+        outputs = []
+        self.log("Retrieved {} nodes...".format(len(nodes)))
+        for node in nodes:
+            outputs.append({
+                "text": node.text,
+                "score": node.score,
+            })
+        if outputs:
+            response = ""
+            for output in outputs:
+                response += "**Score: {}**\n\n{}".format(output["score"], output["text"])
+                if output != outputs[-1]:
+                    response += "\n\n-------\n\n"
+            ctx.set_output(response)
+            ctx.add_doc_meta(self.get_metadata(nodes))
+        return True
+
     def chat(
             self,
             context: BridgeContext,
@@ -225,6 +274,7 @@ class Chat:
         )
 
         if use_index:
+            # CMD: commands are applied to system prompt here
             # index as query engine
             chat_engine = index.as_chat_engine(
                 llm=llm,
@@ -238,6 +288,7 @@ class Chat:
             else:
                 response = chat_engine.chat(query)
         else:
+            # CMD: commands are applied to system prompt here
             # prepare tools (native calls if enabled)
             tools = self.window.core.agents.tools.prepare(context, extra)
 
@@ -245,15 +296,25 @@ class Chat:
             history.insert(0, self.context.add_system(system_prompt))
             history.append(self.context.add_user(query))
             if stream:
-                response = llm.stream_chat_with_tools(
-                    tools=tools,
-                    messages=history,
-                )
+                if hasattr(llm, "stream_chat_with_tools"):
+                    response = llm.stream_chat_with_tools(
+                        tools=tools,
+                        messages=history,
+                    )
+                else:
+                    response = llm.stream_chat(
+                        messages=history,
+                    )
             else:
-                response = llm.chat_with_tools(
-                    tools=tools,
-                    messages=history,
-                )
+                if hasattr(llm, "chat_with_tools"):
+                    response = llm.chat_with_tools(
+                        tools=tools,
+                        messages=history,
+                    )
+                else:
+                    response = llm.chat(
+                        messages=history,
+                    )
 
         # handle response
         if response:
@@ -275,10 +336,9 @@ class Chat:
                     if output is None:
                         output = ""
                     ctx.set_output(output, "")
-                    print("output", output)
                     ctx.add_doc_meta(self.get_metadata(response.source_nodes))  # store metadata
             else:
-                # from LLM
+                # from LLM directly
                 if stream:
                     # tools handled in stream output controller
                     ctx.stream = response  # chunk is in response.delta

pygpt_net/data/config/config.json

@@ -1,7 +1,7 @@
 {
   "__meta__": {
-    "version": "2.4.45",
-    "app.version": "2.4.45",
+    "version": "2.4.46",
+    "app.version": "2.4.46",
     "updated_at": "2024-12-16T00:00:00"
   },
   "access.audio.event.speech": false,
@@ -58,8 +58,13 @@
   "agent.llama.verbose": false,
   "agent.mode": "chat",
   "ai_name": "",
+  "api_azure_version": "2023-07-01-preview",
+  "api_azure_endpoint": "https://<your-resource-name>.openai.azure.com/",
   "api_endpoint": "https://api.openai.com/v1",
   "api_key": "",
+  "api_key_google": "",
+  "api_key_anthropic": "",
+  "api_key_hugging_face": "",
   "api_proxy": "",
   "app.env": [],
   "assistant": "",