pygpt-net 2.6.52__tar.gz → 2.6.54__tar.gz

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/CHANGELOG.md

@@ -1,5 +1,16 @@
 # CHANGELOG
 
+## 2.6.54 (2025-09-18)
+
+- Added: Remote tools (like web search) are now also available in the Chat with Files and Agents (LlamaIndex) modes.
+- Added: Two new plugins: Wolfram Alpha and OpenStreetMap.
+- Fixed: Enabled local file-like schemes in links/images in the markdown-it parser.
+
+## 2.6.53 (2025-09-17)
+
+- Added: An icon to enable/disable the web search remote tool in the icon bar, along with remote web search functionality in OpenRouter (#135).
+- Added: The ability to mute audio in real-time mode via the audio icon.
+
 ## 2.6.52 (2025-09-17)
 
 - Added MCP plugin: Provides access to remote tools via the Model Context Protocol (MCP), including stdio, SSE, and Streamable HTTP transports, with per-server allow/deny filtering, Authorization header support, and a tools cache.

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: pygpt-net
-Version: 2.6.52
+Version: 2.6.54
 Summary: Desktop AI Assistant powered by: OpenAI GPT-5, GPT-4, o1, o3, Gemini, Claude, Grok, DeepSeek, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, MCP, internet access, file handling, command execution and more.
 License: MIT
 Keywords: ai,api,api key,app,assistant,bielik,chat,chatbot,chatgpt,claude,dall-e,deepseek,desktop,gemini,gpt,gpt-3.5,gpt-4,gpt-4-vision,gpt-4o,gpt-5,gpt-oss,gpt3.5,gpt4,grok,langchain,llama-index,llama3,mistral,o1,o3,ollama,openai,presets,py-gpt,py_gpt,pygpt,pyside,qt,text completion,tts,ui,vision,whisper
@@ -119,7 +119,7 @@ Description-Content-Type: text/markdown
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.6.52** | build: **2025-09-17** | Python: **>=3.10, <3.14**
+Release: **2.6.54** | build: **2025-09-18** | Python: **>=3.10, <3.14**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -1450,6 +1450,8 @@ The following plugins are currently available, and model can use them instantly:
 
 - `Mouse and Keyboard` - provides the ability to control the mouse and keyboard by the model.
 
+- `OpenStreetMap` -  Search, geocode, plan routes, and generate static maps using OpenStreetMap services (Nominatim, OSRM, staticmap).
+
 - `Real Time` - automatically appends the current date and time to the system prompt, informing the model about current time.
 
 - `Serial port / USB` - plugin provides commands for reading and sending data to USB ports.
@@ -1474,6 +1476,8 @@ The following plugins are currently available, and model can use them instantly:
 
 - `Wikipedia` - Search Wikipedia for information.
 
+- `Wolfram Alpha` - Compute and solve with Wolfram Alpha: short answers, full JSON pods, math (solve, derivatives, integrals), unit conversions, matrix operations, and plots.
+
 - `X/Twitter` - Interact with tweets and users, manage bookmarks and media, perform likes, retweets, and more.
 
 
@@ -1795,6 +1799,20 @@ Plugin capabilities include:
 
 Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#mouse-and-keyboard
 
+## OpenStreetMap
+
+Provides everyday mapping utilities using OpenStreetMap services:
+
+- Forward and reverse geocoding via Nominatim
+- Search with optional near/bbox filters
+- Routing via OSRM (driving, walking, cycling)
+- Generate openstreetmap.org URL (center/zoom or bbox; optional marker)
+- Utility helpers: open an OSM website URL centered on a point; download a single XYZ tile
+
+Images are saved under `data/openstreetmap/` in the user data directory.
+
+Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#openstreetmap
+
 ## Real Time
 
 This plugin automatically adds the current date and time to each system prompt you send. 
@@ -1918,6 +1936,12 @@ The Wikipedia plugin allows for comprehensive interactions with Wikipedia, inclu
 
 Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#wikipedia
 
+## Wolfram Alpha
+
+Provides computational knowledge via Wolfram Alpha: short answers, full JSON pods, numeric and symbolic math (solve, derivatives, integrals), unit conversions, matrix operations, and plots rendered as images. Images are saved under `data/wolframalpha/` in the user data directory.
+
+Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#wolfram-alpha
+
 ## X/Twitter
 
 The X/Twitter plugin integrates with the X platform, allowing for comprehensive interactions such as tweeting, retweeting, liking, media uploads, and more. This plugin requires OAuth2 authentication and offers various configuration options to manage API interactions effectively.
@@ -3622,6 +3646,17 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.6.54 (2025-09-18)**
+
+- Added: Remote tools (like web search) are now also available in the Chat with Files and Agents (LlamaIndex) modes.
+- Added: Two new plugins: Wolfram Alpha and OpenStreetMap.
+- Fixed: Enabled local file-like schemes in links/images in the markdown-it parser.
+
+**2.6.53 (2025-09-17)**
+
+- Added: An icon to enable/disable the web search remote tool in the icon bar, along with remote web search functionality in OpenRouter (#135).
+- Added: The ability to mute audio in real-time mode via the audio icon.
+
 **2.6.52 (2025-09-17)**
 
 - Added MCP plugin: Provides access to remote tools via the Model Context Protocol (MCP), including stdio, SSE, and Streamable HTTP transports, with per-server allow/deny filtering, Authorization header support, and a tools cache.

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/README.md

@@ -2,7 +2,7 @@
 
 [![pygpt](https://snapcraft.io/pygpt/badge.svg)](https://snapcraft.io/pygpt)
 
-Release: **2.6.52** | build: **2025-09-17** | Python: **>=3.10, <3.14**
+Release: **2.6.54** | build: **2025-09-18** | Python: **>=3.10, <3.14**
 
 > Official website: https://pygpt.net | Documentation: https://pygpt.readthedocs.io
 > 
@@ -1333,6 +1333,8 @@ The following plugins are currently available, and model can use them instantly:
 
 - `Mouse and Keyboard` - provides the ability to control the mouse and keyboard by the model.
 
+- `OpenStreetMap` -  Search, geocode, plan routes, and generate static maps using OpenStreetMap services (Nominatim, OSRM, staticmap).
+
 - `Real Time` - automatically appends the current date and time to the system prompt, informing the model about current time.
 
 - `Serial port / USB` - plugin provides commands for reading and sending data to USB ports.
@@ -1357,6 +1359,8 @@ The following plugins are currently available, and model can use them instantly:
 
 - `Wikipedia` - Search Wikipedia for information.
 
+- `Wolfram Alpha` - Compute and solve with Wolfram Alpha: short answers, full JSON pods, math (solve, derivatives, integrals), unit conversions, matrix operations, and plots.
+
 - `X/Twitter` - Interact with tweets and users, manage bookmarks and media, perform likes, retweets, and more.
 
 
@@ -1678,6 +1682,20 @@ Plugin capabilities include:
 
 Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#mouse-and-keyboard
 
+## OpenStreetMap
+
+Provides everyday mapping utilities using OpenStreetMap services:
+
+- Forward and reverse geocoding via Nominatim
+- Search with optional near/bbox filters
+- Routing via OSRM (driving, walking, cycling)
+- Generate openstreetmap.org URL (center/zoom or bbox; optional marker)
+- Utility helpers: open an OSM website URL centered on a point; download a single XYZ tile
+
+Images are saved under `data/openstreetmap/` in the user data directory.
+
+Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#openstreetmap
+
 ## Real Time
 
 This plugin automatically adds the current date and time to each system prompt you send. 
@@ -1801,6 +1819,12 @@ The Wikipedia plugin allows for comprehensive interactions with Wikipedia, inclu
 
 Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#wikipedia
 
+## Wolfram Alpha
+
+Provides computational knowledge via Wolfram Alpha: short answers, full JSON pods, numeric and symbolic math (solve, derivatives, integrals), unit conversions, matrix operations, and plots rendered as images. Images are saved under `data/wolframalpha/` in the user data directory.
+
+Documentation: https://pygpt.readthedocs.io/en/latest/plugins.html#wolfram-alpha
+
 ## X/Twitter
 
 The X/Twitter plugin integrates with the X platform, allowing for comprehensive interactions such as tweeting, retweeting, liking, media uploads, and more. This plugin requires OAuth2 authentication and offers various configuration options to manage API interactions effectively.
@@ -3505,6 +3529,17 @@ may consume additional tokens that are not displayed in the main window.
 
 ## Recent changes:
 
+**2.6.54 (2025-09-18)**
+
+- Added: Remote tools (like web search) are now also available in the Chat with Files and Agents (LlamaIndex) modes.
+- Added: Two new plugins: Wolfram Alpha and OpenStreetMap.
+- Fixed: Enabled local file-like schemes in links/images in the markdown-it parser.
+
+**2.6.53 (2025-09-17)**
+
+- Added: An icon to enable/disable the web search remote tool in the icon bar, along with remote web search functionality in OpenRouter (#135).
+- Added: The ability to mute audio in real-time mode via the audio icon.
+
 **2.6.52 (2025-09-17)**
 
 - Added MCP plugin: Provides access to remote tools via the Model Context Protocol (MCP), including stdio, SSE, and Streamable HTTP transports, with per-server allow/deny filtering, Authorization header support, and a tools cache.

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pygpt-net"
-version = "2.6.52"
+version = "2.6.54"
 description = "Desktop AI Assistant powered by: OpenAI GPT-5, GPT-4, o1, o3, Gemini, Claude, Grok, DeepSeek, and other models supported by Llama Index, and Ollama. Chatbot, agents, completion, image generation, vision analysis, speech-to-text, plugins, MCP, internet access, file handling, command execution and more."
 authors = ["Marcin Szczyglinski <info@pygpt.net>"]
 license = "MIT"

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/CHANGELOG.txt

@@ -1,3 +1,14 @@
+2.6.54 (2025-09-18)
+
+- Added: Remote tools (like web search) are now also available in the Chat with Files and Agents (LlamaIndex) modes.
+- Added: Two new plugins: Wolfram Alpha and OpenStreetMap.
+- Fixed: Enabled local file-like schemes in links/images in the markdown-it parser.
+
+2.6.53 (2025-09-17)
+
+- Added: An icon to enable/disable the web search remote tool in the icon bar, along with remote web search functionality in OpenRouter (#135).
+- Added: The ability to mute audio in real-time mode via the audio icon.
+
 2.6.52 (2025-09-17)
 
 - Added MCP plugin: Provides access to remote tools via the Model Context Protocol (MCP), including stdio, SSE, and Streamable HTTP transports, with per-server allow/deny filtering, Authorization header support, and a tools cache.

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/__init__.py

@@ -6,15 +6,15 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.09.17 00:00:00                  #
+# Updated Date: 2025.09.18 00:00:00                  #
 # ================================================== #
 
 __author__ = "Marcin Szczygliński"
 __copyright__ = "Copyright 2025, Marcin Szczygliński"
 __credits__ = ["Marcin Szczygliński"]
 __license__ = "MIT"
-__version__ = "2.6.52"
-__build__ = "2025-09-17"
+__version__ = "2.6.54"
+__build__ = "2025-09-18"
 __maintainer__ = "Marcin Szczygliński"
 __github__ = "https://github.com/szczyglis-dev/py-gpt"
 __report__ = "https://github.com/szczyglis-dev/py-gpt/issues"

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/app.py

@@ -97,6 +97,8 @@ from pygpt_net.plugin.server import Plugin as ServerPlugin
 from pygpt_net.plugin.tuya import Plugin as TuyaPlugin
 from pygpt_net.plugin.wikipedia import Plugin as WikipediaPlugin
 from pygpt_net.plugin.mcp import Plugin as MCPPlugin
+from pygpt_net.plugin.wolfram import Plugin as WolframPlugin
+from pygpt_net.plugin.osm import Plugin as OSMPlugin
 
 # agents (Llama-index)
 # from pygpt_net.provider.agents.llama_index.legacy.openai import OpenAIAgent
@@ -427,6 +429,8 @@ def run(**kwargs):
     launcher.add_plugin(TuyaPlugin())
     launcher.add_plugin(WikipediaPlugin())
     launcher.add_plugin(MCPPlugin())
+    launcher.add_plugin(WolframPlugin())
+    launcher.add_plugin(OSMPlugin())
 
     # register custom plugins
     plugins = kwargs.get('plugins', None)

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/audio/audio.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.31 23:00:00                  #
+# Updated Date: 2025.09.17 07:00:00                  #
 # ================================================== #
 
 import os
@@ -36,6 +36,7 @@ class Audio:
             Tab.TAB_CHAT,
             Tab.TAB_TOOL_CALENDAR,
         ]
+        self.muted = False
 
     def setup(self):
         """Setup controller"""
@@ -53,6 +54,26 @@ class Audio:
         if self.window.core.config.get("audio.input.loop", False):
             self.window.ui.nodes['audio.loop'].box.setChecked(True)
 
+    def is_muted(self) -> bool:
+        """
+        Check if audio output is muted
+
+        :return: True if muted
+        """
+        return self.muted
+
+    def set_muted(self, state: bool):
+        """
+        Set audio output muted state
+
+        :param state: True to mute, False to unmute
+        """
+        self.muted = state
+        if state:
+            self.toggle_output_icon(False)
+        else:
+            self.toggle_output_icon(True)
+
     def execute_input_stop(self):
         """Execute input stop (from UI)"""
         self.window.dispatch(Event(Event.AUDIO_INPUT_RECORD_TOGGLE, {

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/chat/chat.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.23 15:00:00                  #
+# Updated Date: 2025.09.17 05:00:00                  #
 # ================================================== #
 
 from typing import Any
@@ -22,6 +22,7 @@ from .files import Files
 from .image import Image
 from .input import Input
 from .output import Output
+from .remote_tools import RemoteTools
 from .render import Render
 from .response import Response
 from .stream import Stream
@@ -46,6 +47,7 @@ class Chat:
         self.image = Image(w)
         self.input = Input(w)
         self.output = Output(w)
+        self.remote_tools = RemoteTools(w)
         self.render = Render(w)
         self.response = Response(w)
         self.stream = Stream(w)
@@ -60,12 +62,14 @@ class Chat:
         """Setup"""
         self.common.setup()
         self.attachment.setup()
+        self.remote_tools.setup()
 
     def reload(self) -> None:
         """Reload"""
         self.common.setup()
         self.render.reload()
         self.attachment.reload()
+        self.remote_tools.setup()
 
     def handle_error(self, err: Any) -> None:
         """

pygpt_net-2.6.54/src/pygpt_net/controller/chat/remote_tools.py

@@ -0,0 +1,116 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# ================================================== #
+# This file is a part of PYGPT package               #
+# Website: https://pygpt.net                         #
+# GitHub:  https://github.com/szczyglis-dev/py-gpt   #
+# MIT License                                        #
+# Created By  : Marcin Szczygliński                  #
+# Updated Date: 2025.09.17 20:00:00                  #
+# ================================================== #
+
+from typing import Union
+
+from pygpt_net.item.model import ModelItem
+
+
+class RemoteTools:
+    def __init__(self, window=None):
+        """
+        Remote tools controller
+
+        :param window: Window instance
+        """
+        self.window = window
+        self.enabled_global = {
+            "web_search": False,
+        }
+
+    def setup(self):
+        """
+        Setup remote tools
+
+        :return: None
+        """
+        cfg_get = self.window.core.config.get
+        self.enabled_global["web_search"] = cfg_get("remote_tools.global.web_search", False)
+        self.update_icons()
+
+    def enabled(self, model: Union[ModelItem, str], tool_name: str) -> bool:
+        """
+        Check if remote tool is enabled
+
+        :param model: ModelItem or model name
+        :param tool_name: Tool name
+        :return: True if enabled, False otherwise
+        """
+        if isinstance(model, str):
+            model = self.window.core.models.get(model)
+        if not model:
+            return False
+        if tool_name == "web_search":
+            return self.is_web(model)
+        return False
+
+    def is_web(self, model: ModelItem) -> bool:
+        """
+        Check if web search is enabled for the given provider
+
+        :param model: ModelItem
+        :return: True if web search is enabled, False otherwise
+        """
+        # at first, check provider-specific config
+        cfg_get = self.window.core.config.get
+        state = False
+        if model.provider == "openai":  # native SDK, responses API
+            state = cfg_get("remote_tools.web_search", False)
+        elif model.provider == "google":  # native SDK
+            state = cfg_get("remote_tools.google.web_search", False)
+        elif model.provider == "anthropic":  # native SDK
+            state = cfg_get("remote_tools.anthropic.web_search", False)
+        elif model.provider == "x_ai":  # native SDK
+            mode = cfg_get("remote_tools.xai.mode", "auto")
+            if mode not in ("auto", "on", "off"):
+                mode = "auto"
+            if mode == "auto" or mode == "on":
+                state = True
+
+        # if not enabled by default or other provider, then use global config
+        if not state:
+            state = self.enabled_global["web_search"]
+
+        return state
+
+    def update_icons(self):
+        """
+        Update remote tools icons in chat tabs
+        """
+        state = self.enabled_global["web_search"]
+        if state:
+            self.window.ui.nodes['icon.remote_tool.web'].set_icon(":/icons/web_on.svg")
+        else:
+            self.window.ui.nodes['icon.remote_tool.web'].set_icon(":/icons/web_off.svg")
+
+    def toggle(self, tool_name: str):
+        """
+        Toggle remote tool (for global toggle button)
+
+        :param tool_name: Tool name
+        """
+        cfg_set = self.window.core.config.set
+
+        # tool: web search
+        if tool_name == "web_search":
+            state = not self.enabled_global["web_search"]
+            self.enabled_global["web_search"] = state
+            cfg_set("remote_tools.global.web_search", state)
+            cfg_set("remote_tools.web_search", state)
+            cfg_set("remote_tools.google.web_search", state)
+            cfg_set("remote_tools.anthropic.web_search", state)
+
+            # xAI has 3 modes: auto, on, off
+            cfg_set("remote_tools.xai.mode", "on" if state else "off")
+
+        # save config
+        self.window.core.config.save()
+        self.update_icons()

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/lang/mapping.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.15 23:00:00                  #
+# Updated Date: 2025.09.17 05:00:00                  #
 # ================================================== #
 
 from typing import Dict
@@ -423,6 +423,7 @@ class Mapping:
         tooltips['icon.video.capture'] = 'icon.video.capture'
         tooltips['icon.audio.output'] = 'icon.audio.output'
         tooltips['icon.audio.input'] = 'icon.audio.input'
+        tooltips['icon.remote_tool.web'] = 'icon.remote_tool.web'
         tooltips['assistant.store.btn.refresh_status'] = 'dialog.assistant.store.btn.refresh_status'
         tooltips['agent.llama.loop.score'] = 'toolbox.agent.llama.loop.score.tooltip'
 

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/mode/mode.py

@@ -6,13 +6,13 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.28 09:00:00                  #
+# Updated Date: 2025.09.17 07:00:00                  #
 # ================================================== #
 
 from pygpt_net.core.events import Event, AppEvent
 from pygpt_net.core.types import (
     MODE_ASSISTANT,
-    MODE_CHAT,
+    MODE_CHAT, MODE_AUDIO,
 )
 from pygpt_net.item.ctx import CtxItem
 from pygpt_net.utils import trans
@@ -92,6 +92,9 @@ class Mode:
                 else:
                     c.assistant.select_current()
 
+            elif mode == MODE_AUDIO:
+                c.audio.set_muted(False) # un-mute and show audio output icon by default
+
             cfg.set('mode', mode)
 
             # reset model and preset at start

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/plugins/plugins.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.09.05 18:00:00                  #
+# Updated Date: 2025.09.17 07:00:00                  #
 # ================================================== #
 
 from typing import List, Dict, Any, Optional
@@ -14,7 +14,7 @@ from typing import List, Dict, Any, Optional
 from PySide6.QtGui import QAction
 
 from pygpt_net.core.types import (
-    MODE_AGENT,
+    MODE_AGENT, MODE_AUDIO,
 )
 from pygpt_net.controller.plugins.presets import Presets
 from pygpt_net.controller.plugins.settings import Settings
@@ -194,20 +194,46 @@ class Plugins:
 
     def toggle(self, id: str):
         """
-        Toggle plugin
+        Toggle plugin (from menu)
 
         :param id: plugin id
         """
         if self.window.core.plugins.is_registered(id):
             if self.is_enabled(id):
                 self.disable(id)
+                if id == "audio_output":
+                    self.window.controller.audio.set_muted(True)
             else:
                 self.enable(id)
+                if id == "audio_output":
+                    self.window.controller.audio.set_muted(False)
 
         self.window.controller.ui.update_tokens()
         self.window.controller.attachment.update()
         self.presets.save_current()
 
+    def toggle_audio_output(self):
+        """
+        Toggle plugin (from icon, audio output only)
+        """
+        id = "audio_output"
+        mode = self.window.core.config.get('mode')
+
+        if mode == MODE_AUDIO:
+            if not self.window.controller.audio.is_muted():
+                self.window.controller.audio.set_muted(True)
+            else:
+                self.window.controller.audio.set_muted(False)
+            return
+
+        if self.window.core.plugins.is_registered(id):
+            if self.is_enabled(id):
+                self.disable(id)
+            else:
+                self.enable(id)
+
+        self.presets.save_current()
+
     def set_by_tab(self, idx: int):
         """
         Set current plugin by tab index

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/realtime/realtime.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.08.31 23:00:00                  #
+# Updated Date: 2025.09.17 07:00:00                  #
 # ================================================== #
 
 from PySide6.QtCore import Slot, QTimer
@@ -64,6 +64,8 @@ class Realtime:
 
         :param event: RealtimeEvent instance
         """
+        is_muted = self.window.controller.audio.is_muted()  # global mute state
+
         # check if mode is supported
         if not self.is_supported() and isinstance(event, RealtimeEvent):
             event.stop = True # stop further propagation
@@ -75,7 +77,7 @@ class Realtime:
         if event.name == RealtimeEvent.RT_OUTPUT_AUDIO_DELTA:
             self.set_idle()
             payload = event.data.get("payload", None)
-            if payload:
+            if payload and not is_muted:  # do not play if muted
                 self.window.core.audio.output.handle_realtime(payload, self.signals)
 
         # audio input chunk: send to the active realtime client
@@ -148,7 +150,10 @@ class Realtime:
 
         # volume change: update volume in audio output handler
         elif event.name == RealtimeEvent.RT_OUTPUT_AUDIO_VOLUME_CHANGED:
-            volume = event.data.get("volume", 1.0)
+            if not is_muted:
+                volume = event.data.get("volume", 1.0)
+            else:
+                volume = 0.0
             self.window.controller.audio.ui.on_output_volume_change(volume)
 
         # error: audio output error

{pygpt_net-2.6.52 → pygpt_net-2.6.54}/src/pygpt_net/controller/ui/mode.py

@@ -6,7 +6,7 @@
 # GitHub:  https://github.com/szczyglis-dev/py-gpt   #
 # MIT License                                        #
 # Created By  : Marcin Szczygliński                  #
-# Updated Date: 2025.09.01 23:00:00                  #
+# Updated Date: 2025.09.17 20:00:00                  #
 # ================================================== #
 
 from pygpt_net.core.types import (
@@ -59,11 +59,17 @@ class Mode:
         is_audio = mode == MODE_AUDIO
 
         if not is_audio:
-            self.window.ui.nodes['audio.auto_turn'].setVisible(False)
-            self.window.ui.nodes["audio.loop"].setVisible(False)
+            ui_nodes['audio.auto_turn'].setVisible(False)
+            ui_nodes["audio.loop"].setVisible(False)
+            if not ctrl.plugins.is_enabled('audio_output'):
+                ctrl.audio.toggle_output_icon(False)
         else:
-            self.window.ui.nodes['audio.auto_turn'].setVisible(True)
-            self.window.ui.nodes["audio.loop"].setVisible(True)
+            ui_nodes['audio.auto_turn'].setVisible(True)
+            ui_nodes["audio.loop"].setVisible(True)
+            if not ctrl.audio.is_muted():
+                ctrl.audio.toggle_output_icon(True)
+            else:
+                ctrl.audio.toggle_output_icon(False)
 
         if not is_assistant:
             ui_nodes['presets.widget'].setVisible(True)
@@ -189,6 +195,12 @@ class Mode:
         show = self.are_attachments(mode)
         ui_tabs['input'].setTabVisible(1, show)
 
+        # remote tools icon visibility
+        if not is_image and not is_completion:
+            ui_nodes['icon.remote_tool.web'].setVisible(True)
+        else:
+            ui_nodes['icon.remote_tool.web'].setVisible(False)
+
         ui_tabs['input'].setTabVisible(2, is_assistant)
         ui_tabs['input'].setTabVisible(3, (not is_assistant) and (not is_image))