PyPI - khoj - Versions diffs - 1.23.2__py3-none-any.whl → 1.23.3__py3-none-any.whl - Mend

khoj 1.23.2py3-none-any.whl → 1.23.3py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (50) hide show

khoj/interface/email/welcome.html CHANGED Viewed

@@ -1,61 +1,90 @@
 <!DOCTYPE html>
-<html>
-    <head>
-        <title>Welcome to Khoj</title>
-    </head>
-<body>
-<body style="font-family: 'Verdana', sans-serif; font-weight: 400; font-style: normal; padding: 0; text-align: left; width: 600px; margin: 20px auto;">
-<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1, user-scalable=no">
-<a class="logo" href="https://khoj.dev" target="_blank" style="text-decoration: none; text-decoration: underline dotted;">
-    <img src="https://khoj.dev/khoj-logo-sideways-500.png" alt="Khoj Logo" style="width: 100px;">
-</a>
-<div class="calls-to-action" style="margin-top: 20px;">
-    <div>
-        <h1 style="color: #333; font-size: large; font-weight: bold; margin: 0; line-height: 1.5; background-color: #fee285; padding: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.5);">Merge AI with your brain</h1>
-        <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">Hi {{name}}! We are psyched to be part of your journey with personal AI. To better help you, we're committed to staying transparent, accessible, and completely open-source.</p>
-        <a class="button" href="https://app.khoj.dev" target="_blank" style="display: block; width: 200px; text-align: center; padding: 10px; margin-top: 20px; color: #333; background-color: #fee285; text-decoration: none; border-radius: 5px; font-weight: bold; transition: background-color 0.3s ease; box-shadow: 6px 6px rgba(0, 0, 0, 1.0); padding: 4px; font-size: large; text-transform: uppercase;">Get Started</a>
-        <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">You're about to get a whole lot more productive.</p>
-        <div style="display: grid; grid-template-columns: 1fr 1fr; grid-gap: 12px; margin-top: 20px;">
-            <div style="border: 1px solid black; border-radius: 8px; padding: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0); margin-top: 20px;">
-                <a href="https://docs.khoj.dev/features/online_search" style="text-decoration: none; text-decoration: underline dotted;">
-                    <h3 style="color: #333; font-size: large; margin: 0; padding: 0; line-height: 2.0; background-color: #b8f1c7; padding: 8px; ">Ditch the search bar</h3>
-                </a>
-                <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">You don't need to click around Google results and sift through information yourself, because Khoj is connected to the internet.</p>
-            </div>
-            <div style="border: 1px solid black; border-radius: 8px; padding: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0); margin-top: 20px;">
-                <a href="https://app.khoj.dev/agents" style="text-decoration: none; text-decoration: underline dotted;">
-                    <h3 style="color: #333; font-size: large; margin: 0; padding: 0; line-height: 2.0; background-color: #b8f1c7; padding: 8px;">Get a village, not just an agent</h3>
-                </a>
-                <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">Khoj can fill the need for more specialized assistance, <a href="https://blog.khoj.dev/posts/using-khoj-for-studying/">such as tutoring</a>, with its curated agents. You get a whole team, always available.</p>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Welcome to Khoj</title>
+</head>
+<body
+    style="font-family: 'Arial', sans-serif; line-height: 1.6; color: #333; max-width: 600px; margin: 0 auto; padding: 20px; background-color: #f5f5f5;">
+    <div
+        style="background-color: #ffffff; border-radius: 10px; box-shadow: 0 0 20px rgba(0, 0, 0, 0.1); padding: 30px;">
+        <a href="https://khoj.dev" target="_blank"
+            style="display: block; text-align: center; margin-bottom: 20px; text-decoration: none;">
+            <img src="https://assets.khoj.dev/khoj_logo.png" alt="Khoj Logo" style="width: 120px;">
+        </a>
+        <h1
+            style="font-size: 24px; color: #2c3e50; margin-bottom: 20px; text-align: center; border-bottom: 2px solid #FFA07A; padding-bottom: 10px;">
+            Merge AI with your brain</h1>
+        <p style="font-size: 16px; color: #333; margin-bottom: 20px;">Hi {{name}}! We are psyched to be part of your
+            journey with personal AI. To better help you, we're committed to staying transparent, accessible, and
+            completely open-source.</p>
+        <a href="https://app.khoj.dev" target="_blank"
+            style="display: block; width: 200px; text-align: center; padding: 10px; margin: 20px auto; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px; font-weight: bold; font-size: 16px; text-transform: uppercase;">Get
+            Started</a>
+        <p style="font-size: 16px; color: #333; margin-bottom: 20px;">You're about to get a whole lot more productive.
+        </p>
+        <a href="https://docs.khoj.dev/features/online_search"
+            style="color: #FFA07A; text-decoration: none; font-weight: bold; font-size: 14px;">
+            <div style="display: grid; grid-template-columns: 1fr 1fr; grid-gap: 20px; margin-bottom: 20px;">
+                <div style="background-color: #f8f9fa; border-left: 4px solid #FFA07A; padding: 15px;">
+                    <h3 style="color: #2c3e50; margin-top: 0; font-size: 18px;">Ditch the search bar</h3>
+                    <p style="font-size: 14px; color: #666; margin-bottom: 0;">You don't need to click around Google
+                        results
+                        and sift through information yourself, because Khoj is connected to the internet.</p>
+                </div>
+        </a>
+        <a href="https://app.khoj.dev/agents"
+            style="color: #FFA07A; text-decoration: none; font-weight: bold; font-size: 14px;">
+            <div style="background-color: #f8f9fa; border-left: 4px solid #FFA07A; padding: 15px;">
+                <h3 style="color: #2c3e50; margin-top: 0; font-size: 18px;">Get a village, not just an agent</h3>
+                <p style="font-size: 14px; color: #666; margin-bottom: 0;">Khoj can fill the need for more specialized
+                    assistance, such as tutoring, with its curated agents. You get a whole team, always available.</p>
             </div>
-            <div style="border: 1px solid black; border-radius: 8px; padding: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0); margin-top: 20px;">
-                <a href="https://docs.khoj.dev/category/clients" style="text-decoration: none; text-decoration: underline dotted;">
-                    <h3 style="color: #333; font-size: large; margin: 0; padding: 0; line-height: 2.0; background-color: #b8f1c7; padding: 8px;">Available where you are</h3>
-                </a>
-                <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">Build on top of your digital brain. Khoj stores whatever data you share with it, so you can get answers from your personal notes and documents in your native language. You can engage from your desktop, Obsidian, WhatsApp, or the web.</p>
+        </a>
+        <a href="https://docs.khoj.dev/category/clients"
+            style="color: #FFA07A; text-decoration: none; font-weight: bold; font-size: 14px;">
+            <div style="background-color: #f8f9fa; border-left: 4px solid #FFA07A; padding: 15px;">
+                <h3 style="color: #2c3e50; margin-top: 0; font-size: 18px;">Activate your data</h3>
+                <p style="font-size: 14px; color: #666; margin-bottom: 0;">Build on top of your digital brain. Khoj
+                    stores whatever data you share with it, so you can get answers from your personal notes and
+                    documents in your native language.</p>
             </div>
-            <div style="border: 1px solid black; border-radius: 8px; padding: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0); margin-top: 20px;">
-                <a href="https://blog.khoj.dev/posts/how-khoj-generates-images/" style="text-decoration: none; text-decoration: underline dotted;">
-                    <h3 style="color: #333; font-size: large; margin: 0; padding: 0; line-height: 2.0; background-color: #b8f1c7; padding: 8px;">Create rich, contextual images</h3>
-                </a>
-                <p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">With your shared data, Khoj can help you create astoundingly personal images depicting scenes of what's important to you.</p>
+        </a>
+        <a href="https://blog.khoj.dev/posts/how-khoj-generates-images/"
+            style="color: #FFA07A; text-decoration: none; font-weight: bold; font-size: 14px;">
+            <div style="background-color: #f8f9fa; border-left: 4px solid #FFA07A; padding: 15px;">
+                <h3 style="color: #2c3e50; margin-top: 0; font-size: 18px;">Create rich, contextual images</h3>
+                <p style="font-size: 14px; color: #666; margin-bottom: 0;">With your shared data, Khoj can help you
+                    create astoundingly personal images depicting scenes of what's important to you.</p>
             </div>
-        </div>
+        </a>
     </div>
-</div>
-<p style="color: #333; font-size: medium; margin-top: 20px; padding: 0; line-height: 1.5;">Like something? Dislike something? Searching for some other magical feature? Our inbox is always open for feedback! Reply to this email and say hi to introduce yourself 👋🏽.</p>
-<p style="color: #333; font-size: large; margin-top: 20px; padding: 0; line-height: 1.5;">- The Khoj Team</p>
-<table style="width: 100%; margin-top: 20px;">
-    <tr>
-        <td style="text-align: center;"><a href="https://docs.khoj.dev" target="_blank" style="padding: 8px; color: #333; background-color: #fee285; border-radius: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0);">Docs</a></td>
-        <td style="text-align: center;"><a href="https://github.com/khoj-ai/khoj" target="_blank" style="padding: 8px; color: #333; background-color: #fee285; border-radius: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0);">GitHub</a></td>
-        <td style="text-align: center;"><a href="https://twitter.com/khoj_ai" target="_blank" style="padding: 8px; color: #333; background-color: #fee285; border-radius: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0);">Twitter</a></td>
-        <td style="text-align: center;"><a href="https://www.linkedin.com/company/khoj-ai" target="_blank" style="padding: 8px; color: #333; background-color: #fee285; border-radius: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0);">LinkedIn</a></td>
-        <td style="text-align: center;"><a href="https://discord.gg/BDgyabRM6e" target="_blank" style="padding: 8px; color: #333; background-color: #fee285; border-radius: 8px; box-shadow: 6px 6px rgba(0, 0, 0, 1.0);">Discord</a></td>
-    </tr>
-</table>
+    <p style="font-size: 16px; color: #333; margin-bottom: 20px;">Like something? Dislike something? Searching for
+        some other magical feature? Our inbox is always open for feedback! Reply to this email and say hi to
+        introduce yourself 👋🏽.</p>
+    <div style="font-size: 18px; font-weight: bold; margin-top: 30px; text-align: right;">- The Khoj Team</div>
+    <div style="margin-top: 30px; text-align: center;">
+        <a href="https://docs.khoj.dev" target="_blank"
+            style="display: inline-block; margin: 0 10px; padding: 8px 15px; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px;">Docs</a>
+        <a href="https://github.com/khoj-ai/khoj" target="_blank"
+            style="display: inline-block; margin: 0 10px; padding: 8px 15px; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px;">GitHub</a>
+        <a href="https://twitter.com/khoj_ai" target="_blank"
+            style="display: inline-block; margin: 0 10px; padding: 8px 15px; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px;">Twitter</a>
+        <a href="https://www.linkedin.com/company/khoj-ai" target="_blank"
+            style="display: inline-block; margin: 0 10px; padding: 8px 15px; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px;">LinkedIn</a>
+        <a href="https://discord.gg/BDgyabRM6e" target="_blank"
+            style="display: inline-block; margin: 0 10px; padding: 8px 15px; background-color: #FFA07A; color: #ffffff; text-decoration: none; border-radius: 5px;">Discord</a>
+    </div>
+    </div>
 </body>
 </html>

khoj/main.py CHANGED Viewed

@@ -131,7 +131,7 @@ def run(should_start_server=True):
     logger.info(f"📦 Initializing DB:\n{db_migrate_output.getvalue().strip()}")
     logger.debug(f"🌍 Initializing Web Client:\n{collectstatic_output.getvalue().strip()}")
-    initialization()
+    initialization(not args.non_interactive)
     # Create app directory, if it doesn't exist
     state.config_file.parent.mkdir(parents=True, exist_ok=True)

khoj/processor/content/images/image_to_entries.py CHANGED Viewed

@@ -4,8 +4,6 @@ import os
 from datetime import datetime
 from typing import Dict, List, Tuple
-from rapidocr_onnxruntime import RapidOCR
 from khoj.database.models import Entry as DbEntry
 from khoj.database.models import KhojUser
 from khoj.processor.content.text_to_entries import TextToEntries
@@ -58,7 +56,6 @@ class ImageToEntries(TextToEntries):
         entry_to_location_map: List[Tuple[str, str]] = []
         for image_file in image_files:
             try:
-                loader = RapidOCR()
                 bytes = image_files[image_file]
                 # write the image to a temporary file
                 timestamp_now = datetime.utcnow().timestamp()
@@ -71,13 +68,18 @@ class ImageToEntries(TextToEntries):
                     bytes = image_files[image_file]
                     f.write(bytes)
                 try:
+                    from rapidocr_onnxruntime import RapidOCR
+                    loader = RapidOCR()
                     image_entries_per_file = ""
                     result, _ = loader(tmp_file)
                     if result:
                         expanded_entries = [text[1] for text in result]
                         image_entries_per_file = " ".join(expanded_entries)
                 except ImportError:
-                    logger.warning(f"Unable to process file: {image_file}. This file will not be indexed.")
+                    logger.warning(
+                        f"Unable to process image or scanned file for text: {image_file}. This file will not be indexed."
+                    )
                     continue
                 entry_to_location_map.append((image_entries_per_file, image_file))
                 entries.extend([image_entries_per_file])

khoj/processor/conversation/utils.py CHANGED Viewed

@@ -18,13 +18,20 @@ from khoj.utils.helpers import is_none_or_empty, merge_dicts
 logger = logging.getLogger(__name__)
 model_to_prompt_size = {
+    # OpenAI Models
     "gpt-3.5-turbo": 12000,
-    "gpt-3.5-turbo-0125": 12000,
-    "gpt-4-0125-preview": 20000,
     "gpt-4-turbo-preview": 20000,
+    "gpt-4o": 20000,
     "gpt-4o-mini": 20000,
     "o1-preview": 20000,
     "o1-mini": 20000,
+    # Google Models
+    "gemini-1.5-flash": 20000,
+    "gemini-1.5-pro": 20000,
+    # Anthropic Models
+    "claude-3-5-sonnet-20240620": 20000,
+    "claude-3-opus-20240229": 20000,
+    # Offline Models
     "TheBloke/Mistral-7B-Instruct-v0.2-GGUF": 3500,
     "NousResearch/Hermes-2-Pro-Mistral-7B-GGUF": 3500,
     "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF": 20000,
@@ -163,7 +170,7 @@ def generate_chatml_messages_with_context(
         if loaded_model:
             max_prompt_size = infer_max_tokens(loaded_model.n_ctx(), model_to_prompt_size.get(model_name, math.inf))
         else:
-            max_prompt_size = model_to_prompt_size.get(model_name, 2000)
+            max_prompt_size = model_to_prompt_size.get(model_name, 10000)
     # Scale lookback turns proportional to max prompt size supported by model
     lookback_turns = max_prompt_size // 750
@@ -291,8 +298,6 @@ def reciprocal_conversation_to_chatml(message_pair):
     return [ChatMessage(content=message, role=role) for message, role in zip(message_pair, ["user", "assistant"])]
-def remove_json_codeblock(response):
+def remove_json_codeblock(response: str):
     """Remove any markdown json codeblock formatting if present. Useful for non schema enforceable models"""
-    if response.startswith("```json") and response.endswith("```"):
-        response = response[7:-3]
-    return response
+    return response.removeprefix("```json").removesuffix("```")

khoj/routers/helpers.py CHANGED Viewed

@@ -632,6 +632,7 @@ async def send_message_to_model_wrapper(
             messages=truncated_messages,
             loaded_model=loaded_model,
             model=chat_model,
+            max_prompt_size=max_tokens,
             streaming=False,
             response_type=response_type,
         )
@@ -721,6 +722,7 @@ def send_message_to_model_wrapper_sync(
             system_message=system_message,
             model_name=chat_model,
             loaded_model=loaded_model,
+            max_prompt_size=max_tokens,
             vision_enabled=vision_available,
             model_type=conversation_config.model_type,
         )
@@ -729,6 +731,7 @@ def send_message_to_model_wrapper_sync(
             messages=truncated_messages,
             loaded_model=loaded_model,
             model=chat_model,
+            max_prompt_size=max_tokens,
             streaming=False,
             response_type=response_type,
         )
@@ -739,6 +742,7 @@ def send_message_to_model_wrapper_sync(
             user_message=message,
             system_message=system_message,
             model_name=chat_model,
+            max_prompt_size=max_tokens,
             vision_enabled=vision_available,
             model_type=conversation_config.model_type,
         )

khoj/utils/cli.py CHANGED Viewed

@@ -50,6 +50,12 @@ def cli(args=None):
         default=False,
         help="Run Khoj in anonymous mode. This does not require any login for connecting users.",
     )
+    parser.add_argument(
+        "--non-interactive",
+        action="store_true",
+        default=False,
+        help="Start Khoj in non-interactive mode. Assumes interactive shell unavailable for config. E.g when run via Docker.",
+    )
     args, remaining_args = parser.parse_known_args(args)

khoj/utils/constants.py CHANGED Viewed

@@ -8,8 +8,15 @@ empty_escape_sequences = "\n|\r|\t| "
 app_env_filepath = "~/.khoj/env"
 telemetry_server = "https://khoj.beta.haletic.com/v1/telemetry"
 content_directory = "~/.khoj/content/"
-default_offline_chat_model = "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF"
-default_online_chat_model = "gpt-4o-mini"
+default_offline_chat_models = [
+    "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF",
+    "bartowski/gemma-2-9b-it-GGUF",
+    "bartowski/gemma-2-2b-it-GGUF",
+    "bartowski/Phi-3.5-mini-instruct-GGUF",
+]
+default_openai_chat_models = ["gpt-4o-mini", "gpt-4o"]
+default_gemini_chat_models = ["gemini-1.5-flash", "gemini-1.5-pro"]
+default_anthropic_chat_models = ["claude-3-5-sonnet-20240620", "claude-3-opus-20240229"]
 empty_config = {
     "search-type": {

khoj/utils/initialization.py CHANGED Viewed

@@ -1,25 +1,37 @@
 import logging
 import os
+from typing import Tuple
 from khoj.database.adapters import ConversationAdapters
 from khoj.database.models import (
     ChatModelOptions,
     KhojUser,
     OpenAIProcessorConversationConfig,
+    ServerChatSettings,
     SpeechToTextModelOptions,
     TextToImageModelConfig,
 )
 from khoj.processor.conversation.utils import model_to_prompt_size, model_to_tokenizer
-from khoj.utils.constants import default_offline_chat_model, default_online_chat_model
+from khoj.utils.constants import (
+    default_anthropic_chat_models,
+    default_gemini_chat_models,
+    default_offline_chat_models,
+    default_openai_chat_models,
+)
 logger = logging.getLogger(__name__)
-def initialization():
+def initialization(interactive: bool = True):
     def _create_admin_user():
         logger.info(
             "👩‍✈️ Setting up admin user. These credentials will allow you to configure your server at /server/admin."
         )
+        if not interactive and (not os.getenv("KHOJ_ADMIN_EMAIL") or not os.getenv("KHOJ_ADMIN_PASSWORD")):
+            logger.error(
+                "🚨 Admin user cannot be created. Please set the KHOJ_ADMIN_EMAIL, KHOJ_ADMIN_PASSWORD environment variables or start server in interactive mode."
+            )
+            exit(1)
         email_addr = os.getenv("KHOJ_ADMIN_EMAIL") or input("Email: ")
         password = os.getenv("KHOJ_ADMIN_PASSWORD") or input("Password: ")
         admin_user = KhojUser.objects.create_superuser(email=email_addr, username=email_addr, password=password)
@@ -27,87 +39,103 @@ def initialization():
     def _create_chat_configuration():
         logger.info(
-            "🗣️  Configure chat models available to your server. You can always update these at /server/admin using the credentials of your admin account"
+            "🗣️ Configure chat models available to your server. You can always update these at /server/admin using your admin account"
         )
-        try:
-            use_offline_model = input("Use offline chat model? (y/n): ")
-            if use_offline_model == "y":
-                logger.info("🗣️ Setting up offline chat model")
-                offline_chat_model = input(
-                    f"Enter the offline chat model you want to use. See HuggingFace for available GGUF models (default: {default_offline_chat_model}): "
-                )
-                if offline_chat_model == "":
-                    ChatModelOptions.objects.create(
-                        chat_model=default_offline_chat_model, model_type=ChatModelOptions.ModelType.OFFLINE
-                    )
-                else:
-                    default_max_tokens = model_to_prompt_size.get(offline_chat_model, 2000)
-                    max_tokens = input(
-                        f"Enter the maximum number of tokens to use for the offline chat model (default {default_max_tokens}):"
-                    )
-                    max_tokens = max_tokens or default_max_tokens
-                    default_tokenizer = model_to_tokenizer.get(
-                        offline_chat_model, "hf-internal-testing/llama-tokenizer"
-                    )
-                    tokenizer = input(
-                        f"Enter the tokenizer to use for the offline chat model (default: {default_tokenizer}):"
-                    )
-                    tokenizer = tokenizer or default_tokenizer
-                    ChatModelOptions.objects.create(
-                        chat_model=offline_chat_model,
-                        model_type=ChatModelOptions.ModelType.OFFLINE,
-                        max_prompt_size=max_tokens,
-                        tokenizer=tokenizer,
-                    )
-        except ModuleNotFoundError as e:
-            logger.warning("Offline models are not supported on this device.")
-        use_openai_model = input("Use OpenAI models? (y/n): ")
-        if use_openai_model == "y":
-            logger.info("🗣️ Setting up your OpenAI configuration")
-            api_key = input("Enter your OpenAI API key: ")
-            OpenAIProcessorConversationConfig.objects.create(api_key=api_key)
-            openai_chat_model = input(
-                f"Enter the OpenAI chat model you want to use (default: {default_online_chat_model}): "
-            )
-            openai_chat_model = openai_chat_model or default_online_chat_model
-            default_max_tokens = model_to_prompt_size.get(openai_chat_model, 2000)
-            max_tokens = input(
-                f"Enter the maximum number of tokens to use for the OpenAI chat model (default: {default_max_tokens}): "
-            )
-            max_tokens = max_tokens or default_max_tokens
-            ChatModelOptions.objects.create(
-                chat_model=openai_chat_model, model_type=ChatModelOptions.ModelType.OPENAI, max_prompt_size=max_tokens
-            )
+        # Set up OpenAI's online chat models
+        openai_configured, openai_provider = _setup_chat_model_provider(
+            ChatModelOptions.ModelType.OPENAI,
+            default_openai_chat_models,
+            default_api_key=os.getenv("OPENAI_API_KEY"),
+            vision_enabled=True,
+            is_offline=False,
+            interactive=interactive,
+        )
+        # Setup OpenAI speech to text model
+        if openai_configured:
             default_speech2text_model = "whisper-1"
-            openai_speech2text_model = input(
-                f"Enter the OpenAI speech to text model you want to use (default: {default_speech2text_model}): "
-            )
-            openai_speech2text_model = openai_speech2text_model or default_speech2text_model
+            if interactive:
+                openai_speech2text_model = input(
+                    f"Enter the OpenAI speech to text model you want to use (default: {default_speech2text_model}): "
+                )
+                openai_speech2text_model = openai_speech2text_model or default_speech2text_model
+            else:
+                openai_speech2text_model = default_speech2text_model
             SpeechToTextModelOptions.objects.create(
                 model_name=openai_speech2text_model, model_type=SpeechToTextModelOptions.ModelType.OPENAI
             )
+        # Setup OpenAI text to image model
+        if openai_configured:
             default_text_to_image_model = "dall-e-3"
-            openai_text_to_image_model = input(
-                f"Enter the OpenAI text to image model you want to use (default: {default_text_to_image_model}): "
-            )
-            openai_speech2text_model = openai_text_to_image_model or default_text_to_image_model
+            if interactive:
+                openai_text_to_image_model = input(
+                    f"Enter the OpenAI text to image model you want to use (default: {default_text_to_image_model}): "
+                )
+                openai_text_to_image_model = openai_text_to_image_model or default_text_to_image_model
+            else:
+                openai_text_to_image_model = default_text_to_image_model
             TextToImageModelConfig.objects.create(
-                model_name=openai_text_to_image_model, model_type=TextToImageModelConfig.ModelType.OPENAI
+                model_name=openai_text_to_image_model,
+                model_type=TextToImageModelConfig.ModelType.OPENAI,
+                openai_config=openai_provider,
             )
-        if use_offline_model == "y" or use_openai_model == "y":
-            logger.info("🗣️  Chat model configuration complete")
+        # Set up Google's Gemini online chat models
+        _setup_chat_model_provider(
+            ChatModelOptions.ModelType.GOOGLE,
+            default_gemini_chat_models,
+            default_api_key=os.getenv("GEMINI_API_KEY"),
+            vision_enabled=False,
+            is_offline=False,
+            interactive=interactive,
+            provider_name="Google Gemini",
+        )
-        use_offline_speech2text_model = input("Use offline speech to text model? (y/n): ")
+        # Set up Anthropic's online chat models
+        _setup_chat_model_provider(
+            ChatModelOptions.ModelType.ANTHROPIC,
+            default_anthropic_chat_models,
+            default_api_key=os.getenv("ANTHROPIC_API_KEY"),
+            vision_enabled=False,
+            is_offline=False,
+            interactive=interactive,
+        )
+        # Set up offline chat models
+        _setup_chat_model_provider(
+            ChatModelOptions.ModelType.OFFLINE,
+            default_offline_chat_models,
+            default_api_key=None,
+            vision_enabled=False,
+            is_offline=True,
+            interactive=interactive,
+        )
+        # Explicitly set default chat model
+        chat_models_configured = ChatModelOptions.objects.count()
+        if chat_models_configured > 0:
+            default_chat_model_name = ChatModelOptions.objects.first().chat_model
+            # If there are multiple chat models, ask the user to choose the default chat model
+            if chat_models_configured > 1 and interactive:
+                user_chat_model_name = input(
+                    f"Enter the default chat model to use (default: {default_chat_model_name}): "
+                )
+            else:
+                user_chat_model_name = None
+            # If the user's choice is valid, set it as the default chat model
+            if user_chat_model_name and ChatModelOptions.objects.filter(chat_model=user_chat_model_name).exists():
+                default_chat_model_name = user_chat_model_name
+            # Create a server chat settings object with the default chat model
+            default_chat_model = ChatModelOptions.objects.filter(chat_model=default_chat_model_name).first()
+            ServerChatSettings.objects.create(chat_default=default_chat_model)
+            logger.info("🗣️ Chat model configuration complete")
+        # Set up offline speech to text model
+        use_offline_speech2text_model = "n" if not interactive else input("Use offline speech to text model? (y/n): ")
         if use_offline_speech2text_model == "y":
             logger.info("🗣️ Setting up offline speech to text model")
             # Delete any existing speech to text model options. There can only be one.
@@ -124,6 +152,64 @@ def initialization():
             logger.info(f"🗣️  Offline speech to text model configured to {offline_speech2text_model}")
+    def _setup_chat_model_provider(
+        model_type: ChatModelOptions.ModelType,
+        default_chat_models: list,
+        default_api_key: str,
+        interactive: bool,
+        vision_enabled: bool = False,
+        is_offline: bool = False,
+        provider_name: str = None,
+    ) -> Tuple[bool, OpenAIProcessorConversationConfig]:
+        supported_vision_models = ["gpt-4o-mini", "gpt-4o"]
+        provider_name = provider_name or model_type.name.capitalize()
+        default_use_model = {True: "y", False: "n"}[default_api_key is not None or is_offline]
+        use_model_provider = (
+            default_use_model if not interactive else input(f"Add {provider_name} chat models? (y/n): ")
+        )
+        if use_model_provider != "y":
+            return False, None
+        logger.info(f"️💬 Setting up your {provider_name} chat configuration")
+        chat_model_provider = None
+        if not is_offline:
+            if interactive:
+                user_api_key = input(f"Enter your {provider_name} API key (default: {default_api_key}): ")
+                api_key = user_api_key if user_api_key != "" else default_api_key
+            else:
+                api_key = default_api_key
+            chat_model_provider = OpenAIProcessorConversationConfig.objects.create(api_key=api_key, name=provider_name)
+        if interactive:
+            chat_model_names = input(
+                f"Enter the {provider_name} chat models you want to use (default: {','.join(default_chat_models)}): "
+            )
+            chat_models = chat_model_names.split(",") if chat_model_names != "" else default_chat_models
+            chat_models = [model.strip() for model in chat_models]
+        else:
+            chat_models = default_chat_models
+        for chat_model in chat_models:
+            default_max_tokens = model_to_prompt_size.get(chat_model)
+            default_tokenizer = model_to_tokenizer.get(chat_model)
+            vision_enabled = vision_enabled and chat_model in supported_vision_models
+            chat_model_options = {
+                "chat_model": chat_model,
+                "model_type": model_type,
+                "max_prompt_size": default_max_tokens,
+                "vision_enabled": vision_enabled,
+                "tokenizer": default_tokenizer,
+                "openai_config": chat_model_provider,
+            }
+            ChatModelOptions.objects.create(**chat_model_options)
+        logger.info(f"🗣️ {provider_name} chat model configuration complete")
+        return True, chat_model_provider
     admin_user = KhojUser.objects.filter(is_staff=True).first()
     if admin_user is None:
         while True:
@@ -139,7 +225,8 @@ def initialization():
             try:
                 _create_chat_configuration()
                 break
-            # Some environments don't support interactive input. We catch the exception and return if that's the case. The admin can still configure their settings from the admin page.
+            # Some environments don't support interactive input. We catch the exception and return if that's the case.
+            # The admin can still configure their settings from the admin page.
             except EOFError:
                 return
             except Exception as e:

{khoj-1.23.2.dist-info → khoj-1.23.3.dist-info}/METADATA RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: khoj
-Version: 1.23.2
+Version: 1.23.3
 Summary: Your Second Brain
 Project-URL: Homepage, https://khoj.dev
 Project-URL: Documentation, https://docs.khoj.dev
@@ -61,7 +61,7 @@ Requires-Dist: pymupdf>=1.23.5
 Requires-Dist: python-multipart>=0.0.7
 Requires-Dist: pytz~=2024.1
 Requires-Dist: pyyaml~=6.0
-Requires-Dist: rapidocr-onnxruntime==1.3.22
+Requires-Dist: rapidocr-onnxruntime==1.3.24
 Requires-Dist: requests>=2.26.0
 Requires-Dist: rich>=13.3.1
 Requires-Dist: schedule==1.1.0

khoj 1.23.2__py3-none-any.whl → 1.23.3__py3-none-any.whl

khoj 1.23.2py3-none-any.whl → 1.23.3py3-none-any.whl