add ai docs
perf3ct
|
After Width: | Height: | Size: 186 KiB |
|
After Width: | Height: | Size: 168 KiB |
|
After Width: | Height: | Size: 43 KiB |
|
After Width: | Height: | Size: 172 KiB |
|
After Width: | Height: | Size: 167 KiB |
|
After Width: | Height: | Size: 237 KiB |
|
After Width: | Height: | Size: 202 KiB |
|
After Width: | Height: | Size: 49 KiB |
|
After Width: | Height: | Size: 80 KiB |
|
After Width: | Height: | Size: 191 KiB |
@ -0,0 +1,15 @@
|
||||
# AI Provider Information
|
||||
Currently, we support the following providers:
|
||||
|
||||
* <a class="reference-link" href="AI%20Provider%20Information/Ollama">Ollama</a>
|
||||
* <a class="reference-link" href="AI%20Provider%20Information/OpenAI.md">OpenAI</a>
|
||||
* <a class="reference-link" href="AI%20Provider%20Information/Anthropic.md">Anthropic</a>
|
||||
* Voyage AI
|
||||
|
||||
To set your preferred chat model, you'll want to enter the provider's name here:
|
||||
|
||||
<figure class="image image_resized" style="width:88.38%;"><img style="aspect-ratio:1884/1267;" src="AI Provider Information_im.png" width="1884" height="1267"></figure>
|
||||
|
||||
And to set your preferred embedding provider:
|
||||
|
||||
<figure class="image image_resized" style="width:93.47%;"><img style="aspect-ratio:1907/1002;" src="1_AI Provider Information_im.png" width="1907" height="1002"></figure>
|
||||
|
After Width: | Height: | Size: 47 KiB |
|
After Width: | Height: | Size: 270 KiB |
|
After Width: | Height: | Size: 24 KiB |
|
After Width: | Height: | Size: 13 KiB |
|
After Width: | Height: | Size: 89 KiB |
@ -0,0 +1,25 @@
|
||||
# Installing Ollama
|
||||
[Ollama](https://ollama.com/) can be installed in a variety of ways, and even runs [within a Docker container](https://hub.docker.com/r/ollama/ollama). Ollama will be noticeably quicker when running on a GPU (Nvidia, AMD, Intel), but it can run on CPU and RAM. To install Ollama without any other prerequisites, you can follow their [installer](https://ollama.com/download):
|
||||
|
||||
<figure class="image image_resized" style="width:50.49%;"><img style="aspect-ratio:785/498;" src="3_Installing Ollama_image.png" width="785" height="498"></figure><figure class="image image_resized" style="width:40.54%;"><img style="aspect-ratio:467/100;" src="Installing Ollama_image.png" width="467" height="100"></figure><figure class="image image_resized" style="width:55.73%;"><img style="aspect-ratio:1296/1011;" src="1_Installing Ollama_image.png" width="1296" height="1011"></figure>
|
||||
|
||||
After their installer completes, if you're on Windows, you should see an entry in the start menu to run it:
|
||||
|
||||
<figure class="image image_resized" style="width:66.12%;"><img style="aspect-ratio:1161/480;" src="2_Installing Ollama_image.png" width="1161" height="480"></figure>
|
||||
|
||||
Also, you should have access to the `ollama` CLI via Powershell or CMD:
|
||||
|
||||
<figure class="image image_resized" style="width:86.09%;"><img style="aspect-ratio:1730/924;" src="5_Installing Ollama_image.png" width="1730" height="924"></figure>
|
||||
|
||||
After Ollama is installed, you can go ahead and `pull` the models you want to use and run. Here's a command to pull my favorite tool-compatible model and embedding model as of April 2025:
|
||||
|
||||
```sh
|
||||
ollama pull llama3.1:8b
|
||||
ollama pull mxbai-embed-large
|
||||
```
|
||||
|
||||
Also, you can make sure it's running by going to [http://localhost:11434](http://localhost:11434) and you should get the following response (port 11434 being the “normal” Ollama port):
|
||||
|
||||
<figure class="image"><img style="aspect-ratio:585/202;" src="4_Installing Ollama_image.png" width="585" height="202"></figure>
|
||||
|
||||
Now that you have Ollama up and running, have a few models pulled, you're ready to go to go ahead and start using Ollama as both a chat provider, and embedding provider!
|
||||
|
After Width: | Height: | Size: 5.3 KiB |
|
After Width: | Height: | Size: 198 KiB |
|
After Width: | Height: | Size: 175 KiB |
|
After Width: | Height: | Size: 186 KiB |
|
After Width: | Height: | Size: 168 KiB |
|
After Width: | Height: | Size: 43 KiB |
|
After Width: | Height: | Size: 172 KiB |
|
After Width: | Height: | Size: 167 KiB |
|
After Width: | Height: | Size: 237 KiB |
|
After Width: | Height: | Size: 202 KiB |
|
After Width: | Height: | Size: 49 KiB |
|
After Width: | Height: | Size: 80 KiB |
|
After Width: | Height: | Size: 191 KiB |
@ -0,0 +1,30 @@
|
||||
<p> </p>
|
||||
<p>Currently, we support the following providers:</p>
|
||||
<ul>
|
||||
<li><a class="reference-link" href="#root/pOsGYCXsbNQG/LMAv4Uy3Wk6J/WkM7gsEUyCXs/_help_7EdTxPADv95W">Ollama</a>
|
||||
</li>
|
||||
<li><a class="reference-link" href="#root/pOsGYCXsbNQG/LMAv4Uy3Wk6J/WkM7gsEUyCXs/_help_ZavFigBX9AwP">OpenAI</a>
|
||||
</li>
|
||||
<li><a class="reference-link" href="#root/pOsGYCXsbNQG/LMAv4Uy3Wk6J/WkM7gsEUyCXs/_help_e0lkirXEiSNc">Anthropic</a>
|
||||
</li>
|
||||
<li>Voyage AI</li>
|
||||
</ul>
|
||||
<p> </p>
|
||||
<p>To set your preferred chat model, you'll want to enter the provider's
|
||||
name here:</p>
|
||||
<figure class="image image_resized" style="width:88.38%;">
|
||||
<img style="aspect-ratio:1884/1267;" src="AI Provider Information_im.png"
|
||||
width="1884" height="1267">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
<p>And to set your preferred embedding provider:</p>
|
||||
<figure class="image image_resized"
|
||||
style="width:93.47%;">
|
||||
<img style="aspect-ratio:1907/1002;" src="1_AI Provider Information_im.png"
|
||||
width="1907" height="1002">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
|
After Width: | Height: | Size: 47 KiB |
|
After Width: | Height: | Size: 270 KiB |
|
After Width: | Height: | Size: 24 KiB |
|
After Width: | Height: | Size: 13 KiB |
|
After Width: | Height: | Size: 89 KiB |
@ -0,0 +1,54 @@
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
<p><a href="https://ollama.com/">Ollama</a> can be installed in a variety
|
||||
of ways, and even runs <a href="https://hub.docker.com/r/ollama/ollama">within a Docker container</a>.
|
||||
Ollama will be noticeably quicker when running on a GPU (Nvidia, AMD, Intel),
|
||||
but it can run on CPU and RAM. To install Ollama without any other prerequisites,
|
||||
you can follow their <a href="https://ollama.com/download">installer</a>:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:50.49%;">
|
||||
<img style="aspect-ratio:785/498;" src="3_Installing Ollama_image.png"
|
||||
width="785" height="498">
|
||||
</figure>
|
||||
<figure class="image image_resized" style="width:40.54%;">
|
||||
<img style="aspect-ratio:467/100;" src="Installing Ollama_image.png" width="467"
|
||||
height="100">
|
||||
</figure>
|
||||
<figure class="image image_resized" style="width:55.73%;">
|
||||
<img style="aspect-ratio:1296/1011;" src="1_Installing Ollama_image.png"
|
||||
width="1296" height="1011">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>After their installer completes, if you're on Windows, you should see
|
||||
an entry in the start menu to run it:</p>
|
||||
<figure class="image image_resized"
|
||||
style="width:66.12%;">
|
||||
<img style="aspect-ratio:1161/480;" src="2_Installing Ollama_image.png"
|
||||
width="1161" height="480">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>Also, you should have access to the <code>ollama</code> CLI via Powershell
|
||||
or CMD:</p>
|
||||
<figure class="image image_resized" style="width:86.09%;">
|
||||
<img style="aspect-ratio:1730/924;" src="5_Installing Ollama_image.png"
|
||||
width="1730" height="924">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>After Ollama is installed, you can go ahead and <code>pull</code> the models
|
||||
you want to use and run. Here's a command to pull my favorite tool-compatible
|
||||
model and embedding model as of April 2025:</p><pre><code class="language-text-x-sh">ollama pull llama3.1:8b
|
||||
ollama pull mxbai-embed-large</code></pre>
|
||||
<p> </p>
|
||||
<p>Also, you can make sure it's running by going to <a href="http://localhost:11434">http://localhost:11434</a> and
|
||||
you should get the following response (port 11434 being the “normal” Ollama
|
||||
port):</p>
|
||||
<p> </p>
|
||||
<figure class="image">
|
||||
<img style="aspect-ratio:585/202;" src="4_Installing Ollama_image.png"
|
||||
width="585" height="202">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>Now that you have Ollama up and running, have a few models pulled, you're
|
||||
ready to go to go ahead and start using Ollama as both a chat provider,
|
||||
and embedding provider!</p>
|
||||
<p> </p>
|
||||
|
After Width: | Height: | Size: 5.3 KiB |
|
After Width: | Height: | Size: 198 KiB |
@ -0,0 +1,177 @@
|
||||
<p> </p>
|
||||
<figure class="image image_resized" style="width:63.68%;">
|
||||
<img style="aspect-ratio:1363/1364;" src="Introduction_image.png" width="1363"
|
||||
height="1364">
|
||||
<figcaption>An example chat with an LLM</figcaption>
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>The AI / LLM features within Trilium Notes are designed to allow you to
|
||||
interact with your Notes in a variety of ways, using as many of the major
|
||||
providers as we can support. </p>
|
||||
<p> </p>
|
||||
<p>In addition to being able to send chats to LLM providers such as OpenAI,
|
||||
Anthropic, and Ollama - we also support agentic tool calling, and embeddings.</p>
|
||||
<p> </p>
|
||||
<p>The quickest way to get started is to navigate to the “AI/LLM” settings:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:74.04%;">
|
||||
<img style="aspect-ratio:1916/1906;" src="5_Introduction_image.png" width="1916"
|
||||
height="1906">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>Enable the feature:</p>
|
||||
<figure class="image image_resized" style="width:82.82%;">
|
||||
<img style="aspect-ratio:1911/997;" src="1_Introduction_image.png" width="1911"
|
||||
height="997">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<h2>Embeddings</h2>
|
||||
<p><strong>Embeddings</strong> are important as it allows us to have an compact
|
||||
AI “summary” (it's not human readable text) of each of your Notes, that
|
||||
we can then perform mathematical functions on (such as cosine similarity)
|
||||
to smartly figure out which Notes to send as context to the LLM when you're
|
||||
chatting, among other useful functions.</p>
|
||||
<p>You will then need to set up the AI “provider” that you wish to use to
|
||||
create the embeddings for your Notes. Currently OpenAI, Voyage AI, and
|
||||
Ollama are supported providers for embedding generation.</p>
|
||||
<p>In the following example, we're going to use our self-hosted Ollama instance
|
||||
to create the embeddings for our Notes. You can see additional documentation
|
||||
about installing your own Ollama locally in <a class="reference-link"
|
||||
href="#root/jdjRLhLV3TtI/LMAv4Uy3Wk6J/7EdTxPADv95W/_help_vvUCN7FDkq7G">Installing Ollama</a>.</p>
|
||||
<p>To see what embedding models Ollama has available, you can check out
|
||||
<a
|
||||
href="https://ollama.com/search?c=embedding">this search</a>on their website, and then <code>pull</code> whichever one
|
||||
you want to try out. As of 4/15/25, my personal favorite is <code>mxbai-embed-large</code>.</p>
|
||||
<p>First, we'll need to select the Ollama provider from the tabs of providers,
|
||||
then we will enter in the Base URL for our Ollama. Since our Ollama is
|
||||
running on our local machine, our Base URL is <code>http://localhost:11434</code>.
|
||||
We will then hit the “refresh” button to have it fetch our models:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:82.28%;">
|
||||
<img style="aspect-ratio:1912/1075;" src="4_Introduction_image.png" width="1912"
|
||||
height="1075">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>When selecting the dropdown for the “Embedding Model”, embedding models
|
||||
should be at the top of the list, separated by regular chat models with
|
||||
a horizontal line, as seen below:</p>
|
||||
<figure class="image image_resized"
|
||||
style="width:61.73%;">
|
||||
<img style="aspect-ratio:1232/959;" src="8_Introduction_image.png" width="1232"
|
||||
height="959">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>After selecting an embedding model, embeddings should automatically begin
|
||||
to be generated by checking the embedding statistics at the top of the
|
||||
“AI/LLM” settings panel:</p>
|
||||
<figure class="image image_resized" style="width:67.06%;">
|
||||
<img style="aspect-ratio:1333/499;" src="7_Introduction_image.png" width="1333"
|
||||
height="499">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>If you don't see any embeddings being created, you will want to scroll
|
||||
to the bottom of the settings, and hit “Recreate All Embeddings”:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:65.69%;">
|
||||
<img style="aspect-ratio:1337/1490;" src="3_Introduction_image.png" width="1337"
|
||||
height="1490">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>Creating the embeddings will take some time, and will be regenerated when
|
||||
a Note is created, updated, or deleted (removed).</p>
|
||||
<p>If for some reason you choose to change your embedding provider, or the
|
||||
model used, you'll need to recreate all embeddings.</p>
|
||||
<p> </p>
|
||||
<p> </p>
|
||||
<h2>Tools</h2>
|
||||
<p>Tools are essentially functions that we provide to the various LLM providers,
|
||||
and then LLMs can respond in a specific format that tells us what tool
|
||||
function and parameters they would like to invoke. We then execute these
|
||||
tools, and provide it as additional context in the Chat conversation. </p>
|
||||
<p> </p>
|
||||
<p>These are the tools that currently exist, and will certainly be updated
|
||||
to be more effectively (and even more to be added!):</p>
|
||||
<ul>
|
||||
<li><code>search_notes</code>
|
||||
<ul>
|
||||
<li>Semantic search</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>keyword_search</code>
|
||||
<ul>
|
||||
<li>Keyword-based search</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>attribute_search</code>
|
||||
<ul>
|
||||
<li>Attribute-specific search</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>search_suggestion</code>
|
||||
<ul>
|
||||
<li>Search syntax helper</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>read_note</code>
|
||||
<ul>
|
||||
<li>Read note content (helps the LLM read Notes)</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>create_note</code>
|
||||
<ul>
|
||||
<li>Create a Note</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>update_note</code>
|
||||
<ul>
|
||||
<li>Update a Note</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>manage_attributes</code>
|
||||
<ul>
|
||||
<li>Manage attributes on a Note</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>manage_relationships</code>
|
||||
<ul>
|
||||
<li>Manage the various relationships between Notes</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>extract_content</code>
|
||||
<ul>
|
||||
<li>Used to smartly extract content from a Note</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><code>calendar_integration</code>
|
||||
<ul>
|
||||
<li>Used to find date notes, create date notes, get the daily note, etc.</li>
|
||||
</ul>
|
||||
</li>
|
||||
</ul>
|
||||
<p> </p>
|
||||
<p>When Tools are executed within your Chat, you'll see output like the following:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:66.88%;">
|
||||
<img style="aspect-ratio:1372/1591;" src="6_Introduction_image.png" width="1372"
|
||||
height="1591">
|
||||
</figure>
|
||||
<p>You don't need to tell the LLM to execute a certain tool, it should “smartly”
|
||||
call tools and automatically execute them as needed.</p>
|
||||
<p> </p>
|
||||
<h2>Overview</h2>
|
||||
<p> </p>
|
||||
<p>Now that you know about embeddings and tools, you can just go ahead and
|
||||
use the “Chat with Notes” button, where you can go ahead and start chatting!:</p>
|
||||
<figure
|
||||
class="image image_resized" style="width:60.77%;">
|
||||
<img style="aspect-ratio:1378/539;" src="2_Introduction_image.png" width="1378"
|
||||
height="539">
|
||||
</figure>
|
||||
<p> </p>
|
||||
<p>If you don't see the “Chat with Notes” button on your side launchbar,
|
||||
you might need to move it from the “Available Launchers” section to the
|
||||
“Visible Launchers” section:</p>
|
||||
<figure class="image image_resized" style="width:69.81%;">
|
||||
<img style="aspect-ratio:1765/1287;" src="9_Introduction_image.png" width="1765"
|
||||
height="1287">
|
||||
</figure>
|
||||
|
After Width: | Height: | Size: 175 KiB |