diff --git a/index.md b/index.md index bf3c15e284eebdf144ff9766bc5fd3b1e95cf853..c5e1bb22b22705d1b1cee4fdc08bba777f1e3d6b 100644 --- a/index.md +++ b/index.md @@ -5,6 +5,12 @@ subtitle: JSC's Inference Infrastructure date: February 04, 2025 --- +# Take the slides with you + +{ width=500px } + +--- + ## Website {width=450px} @@ -378,106 +384,6 @@ date: February 04, 2025 --- -# FUTURE - ---- - - - ---- - -## Vision for the (near) future - ---- - - - ---- - -# Blablador, the brand, as an umbrella for *ALL* inference at JSC - ---- - -## JSC Inference umbrella - -- Blablador as LLM is the first step -- Grow to include other types of models for Science and Industry - ---- - -## Use cases so far - -- LLMs for science (e.g. OpenGPT-X, TrustLLM, CosmoSage) -- [Prithvi-EO-2.0](https://www.nas.nasa.gov/SC24/research/project27.php): Geospatial FM for EO (NASA, IBM, JSC). The 300M and 600M models will be released today -- Terramind: Multi-Modal Geospatial FM for EO (ESA Phi-Lab, JSC, DLR, IBM, KP Labs). Model will be released in spring 2025 -- [Helio](https://arxiv.org/abs/2410.10841) NASA’s FM model for Heliophysics, for understating complex solar phenomena (first study design of the model is available) -- JSC/ESA/DLR's upcomping model [FAST-EO](https://eo4society.esa.int/projects/fast-eo/) -- Health: Radiology with Aachen Uniklinik -- JSC/Cern/ECMWF's [Atmorep](https://www.atmorep.org) -- Open models: - - Pango weather - - Graphcast -- With privacy! - ---- - - - ---- - -## Todo - -- Multi-modality: video, audio, text, images -- Auto-RAG with privacy: - - Easy to do badly. Hard to do securely. -- Everything from the previous slide - ---- - -## Potato - -{ width=350px } - ---- - -# Take the slides with you - -{ width=500px } - ---- - -## Questions? - -{ width=500px } - ---- - -## Extra slides - ---- - - -## LLMOps resource - -A No-BS Database of How Companies Actually Deploy LLMs in Production: 300+ Technical Case Studies, Including Self-Hosted LLMs in [https://www.zenml.io/llmops-database](https://www.zenml.io/llmops-database) - ---- - - -> "I think the complexity of Python package management holds down AI application development more than is widely appreciated. AI faces multiple bottlenecks — we need more GPUs, better algorithms, cleaner data in large quantities. But when I look at the day-to-day work of application builders, there’s one additional bottleneck that I think is underappreciated: The time spent wrestling with version management is an inefficiency I hope we can reduce. " - -Andrew Ng, 28.02.2024 - ---- - -> "Building on top of open source can mean hours wrestling with package dependencies, or sometimes even juggling multiple virtual environments or using multiple versions of Python in one application. This is annoying but manageable for experienced developers, but creates a lot of friction for new AI developers entering our field without a background in computer science or software engineering." - -Andrew Ng, 28.02.2024 - - ---- - - ## API Access - Create a token in the [helmholtz codebase settings - (https://codebase.helmholtz.cloud)](https://codebase.helmholtz.cloud/-/user_settings/personal_access_tokens) with the api scope @@ -529,7 +435,6 @@ curl --header "Authorization: Bearer MY_TOKEN_GOES_HERE" https://api.helmholtz - On Continue, choose to add model, choose Other OpenAI-compatible API - Click in Open Config.json at the end - --- ## Demo: VScode + Continue.dev @@ -576,11 +481,105 @@ curl --header "Authorization: Bearer MY_TOKEN_GOES_HERE" https://api.helmholtz --- + +# FUTURE + +--- + + + +--- + +## Vision for the (near) future + +--- + + + +--- + +# Blablador, the brand, as an umbrella for *ALL* inference at JSC + +--- + +## JSC Inference umbrella + +- Blablador as LLM is the first step +- Grow to include other types of models for Science and Industry + +--- + +## Use cases so far + +- LLMs for science (e.g. OpenGPT-X, TrustLLM, CosmoSage) +- [Prithvi-EO-2.0](https://www.nas.nasa.gov/SC24/research/project27.php): Geospatial FM for EO (NASA, IBM, JSC). The 300M and 600M models will be released today +- Terramind: Multi-Modal Geospatial FM for EO (ESA Phi-Lab, JSC, DLR, IBM, KP Labs). Model will be released in spring 2025 +- [Helio](https://arxiv.org/abs/2410.10841) NASA’s FM model for Heliophysics, for understating complex solar phenomena (first study design of the model is available) +- JSC/ESA/DLR's upcomping model [FAST-EO](https://eo4society.esa.int/projects/fast-eo/) +- Health: Radiology with Aachen Uniklinik +- JSC/Cern/ECMWF's [Atmorep](https://www.atmorep.org) +- Open models: + - Pango weather + - Graphcast +- With privacy! + +--- + + + +--- + +## Todo + +- Multi-modality: video, audio, text, images +- Auto-RAG with privacy: + - Easy to do badly. Hard to do securely. +- Everything from the previous slide + +--- + +## Potato + +{ width=350px } + + +--- + +## Questions? + +{ width=500px } + +--- + +## Extra slides + +--- + + +## LLMOps resource + +A No-BS Database of How Companies Actually Deploy LLMs in Production: 300+ Technical Case Studies, Including Self-Hosted LLMs in [https://www.zenml.io/llmops-database](https://www.zenml.io/llmops-database) + +--- + + +> "I think the complexity of Python package management holds down AI application development more than is widely appreciated. AI faces multiple bottlenecks — we need more GPUs, better algorithms, cleaner data in large quantities. But when I look at the day-to-day work of application builders, there’s one additional bottleneck that I think is underappreciated: The time spent wrestling with version management is an inefficiency I hope we can reduce. " + +Andrew Ng, 28.02.2024 + +--- + +> "Building on top of open source can mean hours wrestling with package dependencies, or sometimes even juggling multiple virtual environments or using multiple versions of Python in one application. This is annoying but manageable for experienced developers, but creates a lot of friction for new AI developers entering our field without a background in computer science or software engineering." + +Andrew Ng, 28.02.2024 + +--- + ## Like the slides? Want to use them? Gitlab link to source code of the slides (needs JUDOOR account) -[https://gitlab.jsc.fz-juelich.de/strube1/2024-12-talk-jsc-colloquium](https://gitlab.jsc.fz-juelich.de/strube1/2024-12-talk-jsc-colloquium) +[https://gitlab.jsc.fz-juelich.de/strube1/2025-02-course-helmholtz-munich](https://gitlab.jsc.fz-juelich.de/strube1/2025-02-course-helmholtz-munich) --- diff --git a/public/images/2025-02-talk-helmholtz-munich.png b/public/images/2025-02-talk-helmholtz-munich.png new file mode 100644 index 0000000000000000000000000000000000000000..0330608ad1d37f6e06986a3fb8809c0d87df57f2 Binary files /dev/null and b/public/images/2025-02-talk-helmholtz-munich.png differ diff --git a/public/images/talk-jsc-colloquium.png b/public/images/talk-jsc-colloquium.png deleted file mode 100644 index 63f51f8048e75fdaa5496ebfb3cbc96932efc285..0000000000000000000000000000000000000000 Binary files a/public/images/talk-jsc-colloquium.png and /dev/null differ diff --git a/public/index.html b/public/index.html index c480de7ece705ff507fd7b218a4adde34d375244..8c26aa3d9eef2023711615efd056183925d3d89a 100644 --- a/public/index.html +++ b/public/index.html @@ -232,6 +232,15 @@ width="550" /></h1> <p class="date">February 04, 2025</p> </section> +<section id="take-the-slides-with-you" class="slide level1"> +<h1>Take the slides with you</h1> +<figure> +<img data-src="images/2025-02-talk-helmholtz-munich.png" width="500" +alt="https://go.fzj.de/2025-02-talk-helmholtz-munich" /> +<figcaption +aria-hidden="true">https://go.fzj.de/2025-02-talk-helmholtz-munich</figcaption> +</figure> +</section> <section class="slide level1"> <h2 id="website">Website</h2> @@ -742,6 +751,110 @@ on their IEK7Cloud</li> <p><a href="https://github.com/haesleinhuepf/bia-bob/blob/main/README.md">https://github.com/haesleinhuepf/bia-bob</a></p> </section> +<section class="slide level1"> + +<h2 id="api-access">API Access</h2> +<ul> +<li class="fragment">Create a token in the <a +href="https://codebase.helmholtz.cloud/-/user_settings/personal_access_tokens">helmholtz +codebase settings - (https://codebase.helmholtz.cloud)</a> with the api +scope</li> +<li class="fragment">Copy the token</li> +<li class="fragment">Go to <a +href="https://api.helmholtz-blablador.fz-juelich.de/docs">https://api.helmholtz-blablador.fz-juelich.de/docs</a></li> +<li class="fragment">Click on Authorize, and paste the token</li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-api">Demo: API</h2> +<p><img data-src="images/fastapi-main.png" /></p> +</section> +<section class="slide level1"> + +<h2 id="demo-api-1">Demo: API</h2> +<p><img data-src="images/fastapi-authorize.png" /></p> +</section> +<section class="slide level1"> + +<h2 id="demo-api-2">Demo: API</h2> +<ul> +<li class="fragment">Go to /v1/models</li> +<li class="fragment">Click on Try it out</li> +<li class="fragment">Click on Execute</li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-api-3">Demo: API</h2> +<p><img data-src="images/fastapi-models.png" /></p> +</section> +<section class="slide level1"> + +<h2 id="demo-curl">Demo: cURL</h2> +<ul> +<li class="fragment"><div class="sourceCode" id="cb1"><pre +class="sourceCode bash"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="ex">curl</span> <span class="at">--header</span> <span class="st">"Authorization: Bearer MY_TOKEN_GOES_HERE"</span> https://api.helmholtz-blablador.fz-juelich.de/v1/models</span></code></pre></div></li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-vscode-continue.dev">Demo: VScode + Continue.dev</h2> +<ul> +<li class="fragment">Yes. It DOES run with Emacs too. Ask your favorite +Emacs expert.</li> +<li class="fragment">Yes, vim too!</li> +<li class="fragment">Add <a +href="https://marketplace.visualstudio.com/items?itemName=Continue.continue">continue.dev</a> +extension to VSCode</li> +<li class="fragment">On Continue, choose to add model, choose Other +OpenAI-compatible API</li> +<li class="fragment">Click in Open Config.json at the end</li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-vscode-continue.dev-1">Demo: VScode + Continue.dev</h2> +<p><img data-src="images/continue-add-model.png" height="600" /></p> +</section> +<section class="slide level1"> + +<h2 id="demo-vscode-continue.dev-2">Demo: VScode + Continue.dev</h2> +<ul> +<li class="fragment"><p>Inside config.json, add at the +<code>"models"</code> section:</p></li> +<li class="fragment"><div class="sourceCode" id="cb2"><pre +class="sourceCode json"><code class="sourceCode json"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span> +<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"model"</span><span class="fu">:</span> <span class="st">"AUTODETECT"</span><span class="fu">,</span></span> +<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"title"</span><span class="fu">:</span> <span class="st">"Blablador"</span><span class="fu">,</span></span> +<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"apiKey"</span><span class="fu">:</span> <span class="st">"glpat-YOURKEYHERE"</span><span class="fu">,</span></span> +<span id="cb2-5"><a href="#cb2-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"apiBase"</span><span class="fu">:</span> <span class="st">"https://api.helmholtz-blablador.fz-juelich.de/v1"</span><span class="fu">,</span></span> +<span id="cb2-6"><a href="#cb2-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"provider"</span><span class="fu">:</span> <span class="st">"openai"</span></span> +<span id="cb2-7"><a href="#cb2-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span></code></pre></div></li> +<li class="fragment"><p>Try with the other models you got from the +API!</p></li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-vscode-continue.dev-3">Demo: VScode + Continue.dev</h2> +<ul> +<li class="fragment">Select some code in a python file</li> +<li class="fragment">Type Control-I (cmd-I on Mac) to edit the code, or +Control-L to “talk” to blablador about this code</li> +<li class="fragment">Ask Blablador to explain this code!</li> +<li class="fragment">Can also fix, add tests, etc</li> +</ul> +</section> +<section class="slide level1"> + +<h2 id="demo-vscode-continue.dev-4">Demo: VScode + Continue.dev</h2> +<p><img data-src="images/continue-ask-code.png" /></p> +</section> +<section id="what-can-you-do-with-it" class="slide level1"> +<h1>What can <em><em>you</em></em> do with it?</h1> +<p><img data-src="images/blablador-uncle-sam.png" /></p> +</section> <section id="future" class="slide level1"> <h1>FUTURE</h1> </section> @@ -829,15 +942,6 @@ href="https://www.atmorep.org">Atmorep</a></li> <h2 id="potato">Potato</h2> <p><img data-src="images/IMG_6561.jpg" width="350" /></p> </section> -<section id="take-the-slides-with-you" class="slide level1"> -<h1>Take the slides with you</h1> -<figure> -<img data-src="images/talk-jsc-colloquium.png" width="500" -alt="https://go.fzj.de/2024-12-jsc-colloquium" /> -<figcaption -aria-hidden="true">https://go.fzj.de/2024-12-jsc-colloquium</figcaption> -</figure> -</section> <section class="slide level1"> <h2 id="questions">Questions?</h2> @@ -886,115 +990,11 @@ in computer science or software engineering.”</p> </section> <section class="slide level1"> -<h2 id="api-access">API Access</h2> -<ul> -<li class="fragment">Create a token in the <a -href="https://codebase.helmholtz.cloud/-/user_settings/personal_access_tokens">helmholtz -codebase settings - (https://codebase.helmholtz.cloud)</a> with the api -scope</li> -<li class="fragment">Copy the token</li> -<li class="fragment">Go to <a -href="https://api.helmholtz-blablador.fz-juelich.de/docs">https://api.helmholtz-blablador.fz-juelich.de/docs</a></li> -<li class="fragment">Click on Authorize, and paste the token</li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-api">Demo: API</h2> -<p><img data-src="images/fastapi-main.png" /></p> -</section> -<section class="slide level1"> - -<h2 id="demo-api-1">Demo: API</h2> -<p><img data-src="images/fastapi-authorize.png" /></p> -</section> -<section class="slide level1"> - -<h2 id="demo-api-2">Demo: API</h2> -<ul> -<li class="fragment">Go to /v1/models</li> -<li class="fragment">Click on Try it out</li> -<li class="fragment">Click on Execute</li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-api-3">Demo: API</h2> -<p><img data-src="images/fastapi-models.png" /></p> -</section> -<section class="slide level1"> - -<h2 id="demo-curl">Demo: cURL</h2> -<ul> -<li class="fragment"><div class="sourceCode" id="cb1"><pre -class="sourceCode bash"><code class="sourceCode bash"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a><span class="ex">curl</span> <span class="at">--header</span> <span class="st">"Authorization: Bearer MY_TOKEN_GOES_HERE"</span> https://api.helmholtz-blablador.fz-juelich.de/v1/models</span></code></pre></div></li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-vscode-continue.dev">Demo: VScode + Continue.dev</h2> -<ul> -<li class="fragment">Yes. It DOES run with Emacs too. Ask your favorite -Emacs expert.</li> -<li class="fragment">Yes, vim too!</li> -<li class="fragment">Add <a -href="https://marketplace.visualstudio.com/items?itemName=Continue.continue">continue.dev</a> -extension to VSCode</li> -<li class="fragment">On Continue, choose to add model, choose Other -OpenAI-compatible API</li> -<li class="fragment">Click in Open Config.json at the end</li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-vscode-continue.dev-1">Demo: VScode + Continue.dev</h2> -<p><img data-src="images/continue-add-model.png" height="600" /></p> -</section> -<section class="slide level1"> - -<h2 id="demo-vscode-continue.dev-2">Demo: VScode + Continue.dev</h2> -<ul> -<li class="fragment"><p>Inside config.json, add at the -<code>"models"</code> section:</p></li> -<li class="fragment"><div class="sourceCode" id="cb2"><pre -class="sourceCode json"><code class="sourceCode json"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a> <span class="fu">{</span></span> -<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a> <span class="dt">"model"</span><span class="fu">:</span> <span class="st">"AUTODETECT"</span><span class="fu">,</span></span> -<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a> <span class="dt">"title"</span><span class="fu">:</span> <span class="st">"Blablador"</span><span class="fu">,</span></span> -<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a> <span class="dt">"apiKey"</span><span class="fu">:</span> <span class="st">"glpat-YOURKEYHERE"</span><span class="fu">,</span></span> -<span id="cb2-5"><a href="#cb2-5" aria-hidden="true" tabindex="-1"></a> <span class="dt">"apiBase"</span><span class="fu">:</span> <span class="st">"https://api.helmholtz-blablador.fz-juelich.de/v1"</span><span class="fu">,</span></span> -<span id="cb2-6"><a href="#cb2-6" aria-hidden="true" tabindex="-1"></a> <span class="dt">"provider"</span><span class="fu">:</span> <span class="st">"openai"</span></span> -<span id="cb2-7"><a href="#cb2-7" aria-hidden="true" tabindex="-1"></a> <span class="fu">}</span></span></code></pre></div></li> -<li class="fragment"><p>Try with the other models you got from the -API!</p></li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-vscode-continue.dev-3">Demo: VScode + Continue.dev</h2> -<ul> -<li class="fragment">Select some code in a python file</li> -<li class="fragment">Type Control-I (cmd-I on Mac) to edit the code, or -Control-L to “talk” to blablador about this code</li> -<li class="fragment">Ask Blablador to explain this code!</li> -<li class="fragment">Can also fix, add tests, etc</li> -</ul> -</section> -<section class="slide level1"> - -<h2 id="demo-vscode-continue.dev-4">Demo: VScode + Continue.dev</h2> -<p><img data-src="images/continue-ask-code.png" /></p> -</section> -<section id="what-can-you-do-with-it" class="slide level1"> -<h1>What can <em><em>you</em></em> do with it?</h1> -<p><img data-src="images/blablador-uncle-sam.png" /></p> -</section> -<section class="slide level1"> - <h2 id="like-the-slides-want-to-use-them">Like the slides? Want to use them?</h2> <p>Gitlab link to source code of the slides (needs JUDOOR account)</p> <p><a -href="https://gitlab.jsc.fz-juelich.de/strube1/2024-12-talk-jsc-colloquium">https://gitlab.jsc.fz-juelich.de/strube1/2024-12-talk-jsc-colloquium</a></p> +href="https://gitlab.jsc.fz-juelich.de/strube1/2025-02-course-helmholtz-munich">https://gitlab.jsc.fz-juelich.de/strube1/2025-02-course-helmholtz-munich</a></p> </section> <section class="slide level1">