Small Language Models

[PLACEHOLDER]

Open source models will continue to grow in popularity. Small Language Models (SLMs) are smaller, faster to train with less compute. They can be used for tackling specific cases while being at a lower cost.

drops-different-size-on-green-leaf-jungle-background

Photo by Tobias Bjørkli via Pexels

SLMs can be more efficient

SLMs are faster in inference speed, and they also require less memory and storage.

SLMs and cost

Small Language models can run on less powerful machines, making them more affordable. This could be ideal for experimentation, startups and/or small size companies.

Here is a short list

Tiny Llama. The 1.1B parameters AI Model, trained on 3T Tokens.
Microsoft’s Phi-2. The 2.7B parameters, trained on 1.4T tokens.
Gemini Nano. The 6B parameters.
Deepseek Coder

Steer for a talent transformation strategy (and avoiding AI fatigue)

There was a debate on whether to feature the term “AI” in the title of this article. Honestly, a key motivation for pursuing the research that led to this post was sparked by the widespread excitement about AI appearing constantly in our LinkedIn feed, to the point of feeling the fatigue, and even a bit disappointed in the algorithm of this, and the others, social media and content curated apps. We soon discovered that there is an entire concept called "AI fatigue", not exactly how we were feeling it, but more about the mixed emotions people in the workforce have regarding the use of AI tools. Photo by Mart Production via Pexels (background updated with AI and Adobe tech) From micro blog posts to video podcasts, lately, most of the tech content we encounter revolves around AI. They often sound or read very similar, usually mentioning the same few top providers. The articles (and social posts... at least the popular ones with paid-campaigns behind it) tend to focus less...

Building MCP with TypeScript

MCP servers are popular these days. We’ve been researching and exploring a few code repos, some where missing modularity, others just not having pieces that we were looking for… therefore we decided to build our own, simple and foundational that could be a starting point for those trying to solve for the similar things we were… and we decided to share it with the community, via our public github. MCP host, server,data sources Before we start. Using Typescript and NodeJS was one of our requirements. This proved somewhat challenging because I don't code as frequently these days due to my leadership responsibilities, and I typically prefer working with C# or Python. Colleagues in my tech community have been working with their teams on some of their MCPs going the Python route. Therefore, I said, “I guess we are trying the other route” 😊. One of our reasons to go with TypeScript was due to the need of the integration with APIs, and based on the research, it seems t...

AI Agents is the new thing to talk about

Tech is evolving faster than ever in this AI era, that it feels every week there is something new to talk about, and what you learn weeks back is no longer relevant, or “that AI tools” already has gone through changes that you need to catch up with in order to stay relevant. Fear not, embrace the challenges and learnings, and find applications for it that are good and ethical for this present, and the hereafter. The new “craze” is AI agents, and for good reason! Image generated with NightCafe In contrast with AI chatbots, an AI agent can execute tasks on your behalf. If you are thinking “ that this could be agents that we leave running independently for many days for a group of deliveries ”… Well then you are correct! Are there risks? Should we talk about trust and accountability? The answer for both is yes. I already hinted at it a couple of paragraphs above, when I wrote “ good and ethical ”. AI (Artificial Intelligence) agents are software that work autonomously,...

Assembling MLOps practice - part 1

In one of our previous articles it was highlighted how DevOps manages the End-to-End application cycle, leveraging agility and automation. CI/CD pipelines, collaboration and transparency, monitoring and automation are part of the list on how DevOps leverages and facilitates agility. What if then we bring those to support ML? That is how MLOps comes to the table and starts making sense! Lego Alike data assembly - Generated with Gemini A big tech corporation, or a startup, nowadays will see how it is becoming a requirement to incorporate AI and Machine learning (ML) in their operations. ML components are key parts of the ecosystem, supporting the solutions provided to clients. As a result, DevOps and MLOps have become part of the "secret sauce" for success. What is MLOps Just to bring the definition of what you probably know (or put together based on the above) MLOps focuses on the life-cycle management of machine learning models. It combines machine learning with traditional ...

Search This Blog