Debanjum
51f3af11b5
Fix Qwen 2.5 14B model source to use Q4_K_M quantized model
...
The official Qwen2.5 14B model doesn't mention standard quantization
suffixes like Q4_K_M, so doesn't work with Khoj
2025-01-19 12:27:35 +07:00
sabaimran
f99bd3f1bc
Merge branch 'master' of github.com:khoj-ai/khoj into features/add-chat-controls
2025-01-17 17:49:08 -08:00
sabaimran
af9e906cb5
Use python3 instead of python when running pip install commands in gh actions
2025-01-17 17:48:42 -08:00
sabaimran
c80e0883ee
Use python3 instead of python for install pip commands in GH action
2025-01-17 17:36:46 -08:00
sabaimran
b8c866014d
Improve instruction description for the agent command description for notes
2025-01-17 17:19:59 -08:00
sabaimran
7481f78f22
Remove unused API request
2025-01-17 17:18:47 -08:00
sabaimran
5aadba20a6
Add backend support for hidden agents (not yet enabled)
2025-01-17 16:46:37 -08:00
sabaimran
2fa212061d
Add a ride hand side bar for chat controls
2025-01-17 16:45:50 -08:00
Tuğhan Belbek
849348e638
Handle additional HTTP redirect status code 308 in scheduled chat requests ( #1088 )
...
Closes #1067
2025-01-16 07:03:43 -08:00
Debanjum
00843f4f24
Release Khoj version 1.34.0
2025-01-16 12:11:28 +07:00
Debanjum
ad27f34c96
Support online search using Google Search API ( #1087 )
...
Add official Google Search API as an online search provider.
We currently support Serper.dev and Searxng as online search providers.
2025-01-16 11:41:59 +07:00
Debanjum
a649b03658
Support online search using Google Search API
2025-01-16 11:39:03 +07:00
Sam Ho
f8f159efac
feat: add turnId handling to chat messages and history
2025-01-16 00:44:16 +00:00
sabaimran
42d4d15346
Merge pull request #1054 from khoj-ai/features/add-support-for-mermaidjs
...
We've been having issues generating diagrams with Excalidraw that are any degree of complexity. By contrast, LLMs are able to handle Mermaid.js syntax a lot better, as it's much more forgiving and has a simpler declarative style. Refer to https://mermaid.js.org/ .
Update so that new diagrams are generated with Mermaid.js, while old diagrams generated with Excalidraw can still be viewed.
2025-01-15 11:55:12 -08:00
Debanjum
e2b2b3415e
Fix handling of inline base64 images by Obsidian, Desktop clients
...
Fix for #1082 pushed down adding the `data:image/webp;base64' prefix
of the base64 images to the server image gen API. But the code on the
Obsidian and Desktop client still add these prefixes.
This change stops the Desktop, Obsidian clients from adding the prefix
as it is being handled by the API now. It should resolve showing
images inline in those clients as well
2025-01-15 23:34:23 +07:00
Debanjum
2e585efd2f
Fix end with newline styling issue in style.css to pass lint checks
2025-01-15 19:43:02 +07:00
Debanjum
ed18c04576
Fix wrapping base64 generated image for inline display
...
Resolves #1082
2025-01-15 19:19:31 +07:00
Debanjum
f8b887cabd
Allow using OpenAI (compatible) API for Speech to Text transcription
2025-01-15 19:19:31 +07:00
Debanjum
182c49b41c
Prefer explicitly configured OpenAI API url, key for image gen model
...
Previously we'd use the general openai client, even if the image
generation model has a different api key and base url set.
This change uses the openai config of the image generation models when
set. Otherwise it fallbacks to use the first openai api provider set
2025-01-15 19:19:31 +07:00
Debanjum
24204873c8
Use same openai base url env var name as the official openai client
...
This eases re-use of the OpenAI API across all openai clients,
including chat, image generation, speech to text.
Resolves #1085
2025-01-15 19:19:30 +07:00
Debanjum
63dd3985b5
Support using Embeddings Model exposed via OpenAI (compatible) API ( #1051 )
...
This change adds the ability to use OpenAI, Azure OpenAI or any embedding model exposed behind an OpenAI compatible API (like Ollama, LiteLLM, vLLM etc.).
Khoj previously only supported HuggingFace embedding models running locally on device or via HuggingFaceW inference API endpoint. This allows using commercial embedding models to index your content with Khoj.
2025-01-15 17:39:54 +07:00
Debanjum
a6bf6803b6
Add docs on how to add, edit search model configs when self-hosting
2025-01-15 17:30:18 +07:00
Debanjum
92a1ec7afc
Do not auto restart khoj docker services by default
...
Let folks who want to add that, add it manually if they want to. It
creates too much noise for folks having trouble with self-host setup
2025-01-15 13:09:50 +07:00
Debanjum
85c537a1de
Set default PORT arg in Dockerfile to default Khoj port, 42110
2025-01-15 13:09:50 +07:00
Debanjum
9355381fac
Catch error in call to data sources, output format selection tool AI
...
Previously if the call to this tool AI failed, the API call would
non-gracefully fail on server. This would leave the client hanging in
a wierd state (e.g with spinner running on web app with no indication
of issue).
Also do not show filters in query debug log lines when no filters in query
2025-01-15 13:09:50 +07:00
Debanjum
24ab8450ba
Handle scenario where read chat stream error is not json on web app
2025-01-15 13:09:50 +07:00
sabaimran
0b775c77d3
Merge branch 'master' of github.com:khoj-ai/khoj into features/add-a-knowledge-base-page
2025-01-13 15:07:59 -08:00
Sam Ho
fc6fab4cce
chore: fix format issue from pre-commit hook - trailing-whitespace and end-of-file-fixer
2025-01-13 20:07:48 +00:00
sabaimran
7f329e7e9d
Fix configuration of name field for chatmodel options during initalization
2025-01-12 22:37:08 -08:00
sabaimran
1a00540ee9
Improve error handling in mermaid chart rendering
2025-01-12 22:36:31 -08:00
Osama Ata
96e3d0a7b9
Fix stale lmstudio documentation to set ai model api via admin panel ( #1075 )
...
Use new name `Ai Model API` instead of `OpenAI Processor Conversation Config`
2025-01-12 03:06:01 -08:00
Yash-1511
27165b3f4a
fix: review suggestions
2025-01-12 15:12:14 +05:30
Debanjum
6bd9f6bb61
Give a shorter, simpler name to github workflow to deploy docs
2025-01-12 10:54:56 +07:00
Sam Ho
93687f141a
feat: do not show delete button on system messages
2025-01-11 17:35:57 +00:00
Sam Ho
a9c180d85f
feat: add delete chat message action to the Obsidian plugin
2025-01-11 17:19:40 +00:00
Debanjum
51a774c993
Add contrast to setting card inputs in dark mode on web app
2025-01-11 14:50:47 +07:00
Debanjum
9e8b8dc5a2
Toggle showing api key on web settings page via a visibility toggle
...
- Background
Access to the clipboard API is disabled by certain browsers in non
localhost http scenarios for security reasons.
So the copy API key button doesn't work when khoj is self-hosted
with authentication enabled at a non localhost domain
- Change
This change enables copying API keys by manual text highlight + copy
if copy button is disabled
Resolves #1070
2025-01-11 14:50:47 +07:00
Debanjum
25c39bd7da
Extract api keys setting card into separate component on web app
2025-01-11 14:50:46 +07:00
sabaimran
c30047e859
Fix Obsidian style.css
2025-01-10 22:18:44 -08:00
sabaimran
da2b89e46a
Merge branch 'master' of github.com:khoj-ai/khoj into features/add-a-knowledge-base-page
2025-01-10 22:18:14 -08:00
sabaimran
f170487338
Fix apostrophe in the add documents modal
2025-01-10 21:58:17 -08:00
sabaimran
be4b091a21
Add new line to styles.css
2025-01-10 21:52:52 -08:00
sabaimran
f398e1eb0c
Add codeblock rendering for the mermaidjs diagram in obsidian
2025-01-10 21:46:39 -08:00
Debanjum
6e955e158b
Use normalized email address for new users
...
Not check email deliverability for now to allow air-gapped usage or
authenticated/multi-user setups with admin managed otp
Closes #1069
2025-01-11 12:28:40 +07:00
sabaimran
c441663394
Merge branch 'master' of github.com:khoj-ai/khoj into features/add-support-for-mermaidjs
2025-01-10 21:25:33 -08:00
sabaimran
85c34a5f0f
Merge pull request #1018 from hjamet/master
...
This PR delivers comprehensive improvements to the Khoj plugin across multiple key areas:
🔍 Search Enhancements:
- Added visual loading indicators during search operations
- Implemented color-coded results to distinguish between vault and external files
- Added abort logic for previous requests to improve performance
- Enhanced search feedback with clear status indicators
- Improved empty state handling
🔄 Synchronization Improvements:
- Added configurable sync interval setting in minutes
- Implemented manual "Sync new changes" command
- Enhanced sync timer management with automatic restart
- Improved notification system for sync operations
📁 Folder Management:
- Added granular folder selection for sync
- Implemented intuitive folder suggestion modal
- Enhanced folder list visualization
💅 UI/UX Improvements:
- Added loading animations and spinners
- Enhanced search results visualization with color coding
- Refined chat interface styling
- Improved overall settings panel organization
🔧 Technical Improvements:
- Refactored search and synchronization logic
- Implemented proper request cancellation
- Enhanced error handling and user feedback
- Improved code organization and maintainability
2025-01-10 21:24:12 -08:00
sabaimran
57545c1485
Fix the migration script to delete orphaned fileobjects
...
- Remove knowledge page from the sidebar
- Improve speed and rendering of the documents in the search page
2025-01-10 21:06:48 -08:00
sabaimran
d77984f9d1
Remove separate knowledge base file - consolidated in the search page
2025-01-10 18:57:38 -08:00
sabaimran
f2c6ce2435
Improve rendering of the file objects and sort files by updated_date
2025-01-10 18:18:15 -08:00
Patrick Jackson
6e0c767ff0
Use the configured OpenAI Base URL for Automations ( #1065 )
...
This change makes Automations (and possibly other entrypoints) use the configured OpenAI-compatible server if that has been set. Without this change it tries to use the hardcoded OpenAI provider.
2025-01-10 17:17:51 -08:00
