feat: sso login, demo mode & new mindmap support (#644) bump:minor

* fix: update .env.example

* feat: add SSO login

* fix: update flowsetting

* fix: add requirement

* fix: refine UI

* fix: update group id-based operation

* fix: improve citation logics

* fix: UI enhancement

* fix: user_id to string in models

* fix: improve chat suggestion UI and flow

* fix: improve group id handling

* fix: improve chat suggestion

* fix: secure download for single file

* fix: file limiting in docstore

* fix: improve chat suggestion logics & language conform

* feat: add markmap and select text to highlight function

* fix: update Dockerfile

* fix: user id auto generate

* fix: default user id

* feat: add demo mode

* fix: update flowsetting

* fix: revise default params for demo

* feat: sso_app alternative

* feat: sso login demo

* feat: demo specific customization

* feat: add login using API key

* fix: disable key-based login

* fix: optimize duplicate upload

* fix: gradio routing

* fix: disable arm build for demo

* fix: revise full-text search js logic

* feat: add rate limit

* fix: update Dockerfile with new launch script

* fix: update Dockerfile

* fix: update Dockerignore

* fix: update ratelimit logic

* fix: user_id in user management page

* fix: rename conv logic

* feat: update demo hint

* fix: minor fix

* fix: highlight on long PDF load

* feat: add HF paper list

* fix: update HF papers load logic

* feat: fly config

* fix: update fly config

* fix: update paper list pull api

* fix: minor update root routing

* fix: minor update root routing

* fix: simplify login flow & paper list UI

* feat: add paper recommendation

* fix: update Dockerfile

* fix: update Dockerfile

* fix: update default model

* feat: add long context Ollama through LCOllama

* feat: espose Gradio share to env

* fix: revert customized changes

* fix: list group at app load

* fix: relocate share conv button

* fix: update launch script

* fix: update Docker CI

* feat: add Ollama model selection at first setup

* docs: update README

Author: taprosoft

.dockerignore

@@ -11,3 +11,5 @@ env/
 README.md
 *.zip
 *.sh
+
+!/launch.sh

.env.example

@@ -3,8 +3,8 @@
 # settings for OpenAI
 OPENAI_API_BASE=https://api.openai.com/v1
 OPENAI_API_KEY=<YOUR_OPENAI_KEY>
-OPENAI_CHAT_MODEL=gpt-3.5-turbo
-OPENAI_EMBEDDINGS_MODEL=text-embedding-ada-002
+OPENAI_CHAT_MODEL=gpt-4o-mini
+OPENAI_EMBEDDINGS_MODEL=text-embedding-3-large
 
 # settings for Azure OpenAI
 AZURE_OPENAI_ENDPOINT=
@@ -17,10 +17,8 @@ AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT=text-embedding-ada-002
 COHERE_API_KEY=<COHERE_API_KEY>
 
 # settings for local models
-LOCAL_MODEL=llama3.1:8b
+LOCAL_MODEL=qwen2.5:7b
 LOCAL_MODEL_EMBEDDINGS=nomic-embed-text
-LOCAL_EMBEDDING_MODEL_DIM = 768
-LOCAL_EMBEDDING_MODEL_MAX_TOKENS = 8192
 
 # settings for GraphRAG
 GRAPHRAG_API_KEY=<YOUR_OPENAI_KEY>

.github/workflows/build-push-docker.yaml

@@ -28,6 +28,7 @@ jobs:
         target:
           - lite
           - full
+          - ollama
     steps:
       - name: Free Disk Space (Ubuntu)
         uses: jlumbroso/free-disk-space@main

.github/workflows/fly-deploy.yml

@@ -0,0 +1,18 @@
+# See https://fly.io/docs/app-guides/continuous-deployment-with-github-actions/
+
+name: Fly Deploy
+on:
+  push:
+    branches:
+      - main
+jobs:
+  deploy:
+    name: Deploy app
+    runs-on: ubuntu-latest
+    concurrency: deploy-group # optional: ensure only one action runs at a time
+    steps:
+      - uses: actions/checkout@v4
+      - uses: superfly/flyctl-actions/setup-flyctl@master
+      - run: flyctl deploy --remote-only
+        env:
+          FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }}

.pre-commit-config.yaml

@@ -57,6 +57,7 @@ repos:
             "types-requests",
             "sqlmodel",
             "types-Markdown",
+            "types-cachetools",
             types-tzlocal,
           ]
         args: ["--check-untyped-defs", "--ignore-missing-imports"]

Dockerfile

@@ -35,6 +35,7 @@ RUN bash scripts/download_pdfjs.sh $PDFJS_PREBUILT_DIR
 
 # Copy contents
 COPY . /app
+COPY launch.sh /app/launch.sh
 COPY .env.example /app/.env
 
 # Install pip packages
@@ -54,7 +55,7 @@ RUN apt-get autoremove \
     && rm -rf /var/lib/apt/lists/* \
     && rm -rf ~/.cache
 
-CMD ["python", "app.py"]
+ENTRYPOINT ["sh", "/app/launch.sh"]
 
 # Full version
 FROM lite AS full
@@ -97,7 +98,17 @@ RUN apt-get autoremove \
     && rm -rf /var/lib/apt/lists/* \
     && rm -rf ~/.cache
 
-# Download nltk packages as required for unstructured
-# RUN python -c "from unstructured.nlp.tokenize import _download_nltk_packages_if_not_present; _download_nltk_packages_if_not_present()"
+ENTRYPOINT ["sh", "/app/launch.sh"]
 
-CMD ["python", "app.py"]
+# Ollama-bundled version
+FROM full AS ollama
+
+# Install ollama
+RUN --mount=type=ssh  \
+    --mount=type=cache,target=/root/.cache/pip  \
+    curl -fsSL https://ollama.com/install.sh | sh
+
+# RUN nohup bash -c "ollama serve &" && sleep 4 && ollama pull qwen2.5:7b
+RUN nohup bash -c "ollama serve &" && sleep 4 && ollama pull nomic-embed-text
+
+ENTRYPOINT ["sh", "/app/launch.sh"]

README.md

@@ -96,18 +96,7 @@ documents and developers who want to build their own RAG pipeline.
 
 ### With Docker (recommended)
 
-1. We support both `lite` & `full` version of Docker images. With `full`, the extra packages of `unstructured` will be installed as well, it can support additional file types (`.doc`, `.docx`, ...) but the cost is larger docker image size. For most users, the `lite` image should work well in most cases.
-
-   - To use the `lite` version.
-
-     ```bash
-     docker run \
-     -e GRADIO_SERVER_NAME=0.0.0.0 \
-     -e GRADIO_SERVER_PORT=7860 \
-     -v ./ktem_app_data:/app/ktem_app_data \
-     -p 7860:7860 -it --rm \
-     ghcr.io/cinnamon/kotaemon:main-lite
-     ```
+1. We support both `lite` & `full` version of Docker images. With `full` version, the extra packages of `unstructured` will be installed, which can support additional file types (`.doc`, `.docx`, ...) but the cost is larger docker image size. For most users, the `lite` image should work well in most cases.
 
    - To use the `full` version.
 
@@ -124,9 +113,16 @@ documents and developers who want to build their own RAG pipeline.
 
      ```bash
      # change image name to
-     ghcr.io/cinnamon/kotaemon:feat-ollama_docker-full
+     docker run <...> ghcr.io/cinnamon/kotaemon:main-ollama
      ```
 
+   - To use the `lite` version.
+
+   ```bash
+    # change image name to
+    docker run <...> ghcr.io/cinnamon/kotaemon:main-lite
+   ```
+
 2. We currently support and test two platforms: `linux/amd64` and `linux/arm64` (for newer Mac). You can specify the platform by passing `--platform` in the `docker run` command. For example:
 
    ```bash

app.py

@@ -3,6 +3,7 @@
 from theflow.settings import settings as flowsettings
 
 KH_APP_DATA_DIR = getattr(flowsettings, "KH_APP_DATA_DIR", ".")
+KH_GRADIO_SHARE = getattr(flowsettings, "KH_GRADIO_SHARE", False)
 GRADIO_TEMP_DIR = os.getenv("GRADIO_TEMP_DIR", None)
 # override GRADIO_TEMP_DIR if it's not set
 if GRADIO_TEMP_DIR is None:
@@ -21,4 +22,5 @@
         "libs/ktem/ktem/assets",
         GRADIO_TEMP_DIR,
     ],
+    share=KH_GRADIO_SHARE,
 )

docs/about.md

@@ -4,8 +4,8 @@ An open-source tool for chatting with your documents. Built with both end users
 developers in mind.
 
 [Source Code](https://github.com/Cinnamon/kotaemon) |
-[Live Demo](https://huggingface.co/spaces/cin-model/kotaemon-demo)
+[HF Space](https://huggingface.co/spaces/cin-model/kotaemon-demo)
 
-[User Guide](https://cinnamon.github.io/kotaemon/) |
+[Installation Guide](https://cinnamon.github.io/kotaemon/) |
 [Developer Guide](https://cinnamon.github.io/kotaemon/development/) |
 [Feedback](https://github.com/Cinnamon/kotaemon/issues)

docs/online_install.md

@@ -1,7 +1,7 @@
 ## Installation (Online HuggingFace Space)
 
 1. Go to [HF kotaemon_template](https://huggingface.co/spaces/cin-model/kotaemon_template).
-2. Use Duplicate function to create your own space.
+2. Use Duplicate function to create your own space. Or use this [direct link](https://huggingface.co/spaces/cin-model/kotaemon_template?duplicate=true).
    ![Duplicate space](https://raw.githubusercontent.com/Cinnamon/kotaemon/main/docs/images/duplicate_space.png)
    ![Change space params](https://raw.githubusercontent.com/Cinnamon/kotaemon/main/docs/images/change_space_params.png)
 3. Wait for the build to complete and start up (apprx 10 mins).