Commit graph

4083 commits

Author SHA1 Message Date
Debanjum
b0abec39d5 Use chat model name var name consistently and type openai chat utils 2024-12-12 14:20:34 -08:00
Debanjum
4915be0301 Fix initializing chat model names parameter after field rename in 2024-12-12 14:20:33 -08:00
Debanjum
9eb863e964 Restart code sandbox on crash in eval github workflow 2024-12-12 11:28:54 -08:00
Debanjum
01bc6d35dc
Rename Chat Model Options table to Chat Model as short & readable ()
- Previous was incorrectly plural but was defining only a single model
- Rename chat model table field to name
- Update documentation
- Update references functions and variables to match new name
2024-12-12 11:24:16 -08:00
aditya218
9be26e1bd2
Fix documentation to point to local environment image_generation.md ()
Fix documentation to point to local environment.
2024-12-11 16:12:25 -08:00
Debanjum
fe09df66b7 Make code sandbox container url accessible to Khoj container in docker compose 2024-12-11 01:14:26 -08:00
Debanjum
59008ae90e Use buildx to create multi platform docker image 2024-12-11 00:21:29 -08:00
Debanjum
ec797bc6b8 Build docker imgs on native arch runners to avoid manifest list error
This also avoids the need to use --amend and annotate steps when
creating the multi-arch docker images
2024-12-10 23:16:36 -08:00
Debanjum
5f7b13df2d Fix new docker tags in workflow to not include forward slashes 2024-12-10 22:55:33 -08:00
Debanjum
ba6237b5c0 Fix to create multi-arch builds. Stop docker image overwrites in workflow 2024-12-10 21:08:17 -08:00
sabaimran
44ede26e67 Temporarily disable cloud arm builds while we disambiguate the build issues 2024-12-10 20:00:59 -08:00
Debanjum
33a5efaf4b Fix undefined variable exception during openai provider setup on init
Resolves 
2024-12-10 19:54:00 -08:00
sabaimran
e43341fdcc Release Khoj version 1.31.0 2024-12-10 19:41:31 -08:00
Debanjum
a757ecfd2b Put the generated assets message after the user query and fix prompt 2024-12-10 19:40:13 -08:00
Debanjum
40e4f2ec2e Reduce clutter in chat message ux on Obsidian
- Move khoj message border to left like in web ui
- Remove user, sender emojis and name
- Ensure conversation title always at top of chat sessions view,
  even if no chat sessions loaded yet, instead of causing layout shift
  on chat sessions load
2024-12-10 19:34:17 -08:00
sabaimran
1ec1eff57e Improve mobile header, reduce title bar padding and add conv title 2024-12-10 19:03:57 -08:00
sabaimran
321eeeaed7 Fix setting title of shared conversation, move shared button into the title pane 2024-12-10 18:19:46 -08:00
sabaimran
d7e5a76ace Add an icon in the input bar for research mode 2024-12-10 17:49:25 -08:00
sabaimran
01d000e570
Merge pull request from khoj-ai/features/improve-multiple-output-mode-generation
Improve handling of multiple output modes

- Use the generated descriptions / inferred queries to supply context to the model about what it's created and give a richer response
- Stop sending the generated image in user message. This seemed to be confusing the model more than helping.
- Collect generated assets in a structured objects to provide model context. This seems to help it follow instructions and separate responsibility better
- Also, rename the open ai converse method to converse_openai to follow patterns with other providers
2024-12-10 17:06:19 -08:00
sabaimran
2bb14c55a6 Merge branch 'master' of github.com:khoj-ai/khoj into features/improve-multiple-output-mode-generation 2024-12-10 16:56:36 -08:00
sabaimran
6c8007e23b Improve handling of multiple output modes
- Use the generated descriptions / inferred queries to supply context to the model about what it's created and give a richer response
- Stop sending the generated image in user message. This seemed to be confusing the model more than helping.
- Also, rename the open ai converse method to converse_openai to follow patterns with other providers
2024-12-10 16:54:21 -08:00
Debanjum
4bc5c1357a Upgrade server, documentation dependencies. Spell fix docker-compose.yml 2024-12-10 15:47:47 -08:00
Debanjum
f8957e52bf Keep chatml message content simple for wider compat unless attachments
This allows for wider compatibility with chat models and openai proxy
ai model apis that expect message content to be string format, not
objects.
2024-12-10 00:10:56 -08:00
sabaimran
4b4e0e20d4 Make the version number a badge, rather than an independent item in the nav dropdown 2024-12-09 14:45:26 -08:00
sabaimran
eb36492ba5 Update handling of images when included in the chat history with assistant message 2024-12-08 21:46:07 -08:00
Debanjum
b660c494bc Use recognizable DB model names to ease selection UX on Admin Panel
Previously id were used (by default) for model display strings.
This made it hard to select chat model options, server chat settings
etc. in the admin panel dropdowns.

This change uses more recognizable names for the DB objects to ease
selection in dropdowns and display in general on the admin panel.
2024-12-08 20:34:50 -08:00
Debanjum
d10dc9cfe1 Inform code tool AI only limited python packages are available to it
Reduce code tool failing with module not found errors
2024-12-08 20:34:50 -08:00
Debanjum
3fd8614a4b Only auto load available chat models from Ollama provider for now
Allowing models from any openai proxy service makes it too unwieldy.
And a bunch of them do not even support this endpoint.
2024-12-08 20:34:50 -08:00
sabaimran
2c934162d3 Add a data filter for privacy_level of agents 2024-12-08 19:55:57 -08:00
sabaimran
3b9f4c4356 Correct negative for running prod image locally 2024-12-08 19:55:35 -08:00
Debanjum
9dd3782f5c
Rename OpenAIProcessorConversationConfig DB model to more apt AiModelApi ()
* Rename OpenAIProcessorConversationConfig to more apt AiModelAPI

The DB model name had drifted from what it is being used for,
a general chat api provider that supports other chat api providers like
anthropic and google chat models apart from openai based chat models.

This change renames the DB model and updates the docs to remove this
confusion.

Using Ai Model Api we catch most use-cases including chat, stt, image generation etc.
2024-12-08 18:02:29 -08:00
sabaimran
df66fb23ab Centralize definition of the content security policy and add in-app chat
- in-app chat is meant for support requests and currently is only in the settings page, where users are most likely to go if confused IMO
2024-12-08 17:57:27 -08:00
sabaimran
0b87c13f8d Add khoj_version to the settings menu 2024-12-08 17:55:56 -08:00
sabaimran
47a087c73b Fix chatwoot import issue by checking whether we're in an execution environment before loading the script 2024-12-08 17:16:20 -08:00
sabaimran
66f59c8d41 Add Chatwoot to documentation
See repo: https://github.com/chatwoot/chatwoot
2024-12-08 16:51:43 -08:00
sabaimran
6ed051d631
Merge pull request from khoj-ai/features/update-desktop-app
Simplify the desktop app

- Make the desktop app mainly a file-syncing client for users who have lots of documents that they need to share with Khoj. This is because the web app provides a fairly robust chat client which can be used by anyone on their computer.
- The chat client in the desktop app had significantly drifted from our current brand / them, and didn't provide enough value add to update. Later, we will make it easier to install the existing web app as a desktop PWA.
2024-12-08 15:05:35 -08:00
sabaimran
05b3911080 Update some button titles and add descriptions for clarity 2024-12-08 14:29:12 -08:00
sabaimran
b78b92d6a0 Merge branch 'master' of github.com:khoj-ai/khoj into features/update-desktop-app 2024-12-08 14:20:20 -08:00
sabaimran
e3789aef49
Merge pull request from khoj-ai/features/allow-multi-outputs-in-chat
Currently, Khoj has terminal states with respect to what assets it outputs. We limit it to image, text, and excalidraw diagram. This limitation is unnecessary and provides undue constraints for creating more dynamic, diverse experiences. For instance, we may want the chat view to morph for document editing or generation, in which case having limited output modes would be a detriment.

This change allows us to emit generated assets and then continue on to more text generation in final response. It forces text response for all messages. It adds a new stream event, GENERATED_ASSETS, which holds content that the AI is emitting in response to the query.
2024-12-08 14:19:05 -08:00
sabaimran
a2251f01eb Make result optional for code context, relevant when code execution was unsuccessful 2024-12-08 13:27:33 -08:00
sabaimran
9c403d24e1 Fix reference to directory in the eval workflow for starting terrarium 2024-12-08 13:03:05 -08:00
sabaimran
7cd2855146 Make attributes optional in the knowledge graph model 2024-12-08 12:23:17 -08:00
sabaimran
2af687d1c5 Allow snippetHighlighted to also be nullable 2024-12-08 11:51:24 -08:00
sabaimran
efa23a8ad8 Update validation requirements for online searches 2024-12-08 11:30:17 -08:00
sabaimran
6940c6379b Add sudo when running installations in order to install relevant packages
add --legacy-peer-deps temporarily to see if it helps mitigate the issue
2024-12-08 11:11:13 -08:00
sabaimran
4c4b7120c6 Use Khoj terrarium fork instead of building from official Cohere repo 2024-12-08 11:06:33 -08:00
sabaimran
a138845fea Merge branch 'master' of github.com:khoj-ai/khoj into features/allow-multi-outputs-in-chat 2024-12-08 10:57:16 -08:00
sabaimran
19832a3ed0 Add note to uncomment line when using the prod image 2024-12-05 18:16:01 -08:00
sabaimran
110c64ba27 Update the desktop instructions 2024-12-05 17:39:16 -08:00
Debanjum
65c5b163c9 Add khoj_lantern svg to web public assets for use by new admin panel 2024-12-05 10:57:18 -08:00