Forem: Richa

Explore a clear, real-world case of what happens when you run a local AI model. A classic example of hardware strain during model inference

Richa — Fri, 21 Nov 2025 15:49:47 +0000

When a “Small” AI Model Pushes Your Hardware to Its Limits - DEV Community

While building my ConversaAI web app, I started experimenting with a local AI model using Ollama,...

dev.to

When a “Small” AI Model Pushes Your Hardware to Its Limits

Richa — Fri, 21 Nov 2025 15:48:37 +0000

While building my ConversaAI web app, I started experimenting with a local AI model using Ollama, running the Gemma 3 (1B) model, a "lightweight" 815 MB model.
According to the documentation, a system with 8 GB RAM can handle models up to 7B parameters, so I expected a smooth run since my system has:
• CPU: Intel i7-8650U @ 1.9 GHz
• RAM: 16 GB
• OS: Windows 11 Pro
• GPU: NVIDIA MX130

But in practice… it’s a different story.

Every time model started generating a longer response,
• CPU jumped from 10 % to 60 %
• GPU (NVIDIA MX130) hit 95% utilization.
• Memory climbed to 8.5 GB / 16 GB
• Fans roared like a jet engine. 🌀
…and then the model simply stopped generating mid-response. No error. No crash message.
A few seconds later, the fan also slowed down again.

👇 Here's the before-and-during-generating-longer-response comparison of hardware utilization.

At first, I thought something broke.

But the real reason was far more interesting.

🧠 What Was Really Happening

Even at 1B parameters, the model performs billions of operations per token.

Every generated word triggers massive matrix multiplications. The GPU handles the heavy computation, while the CPU manages data flow, scheduling, and memory movement. Meanwhile, RAM and VRAM temporarily hold weights, activations, and intermediate states.

In simple terms:
• The GPU does intense math.
• The CPU coordinates everything.
• RAM/VRAM holds model weights and activations.
• The fan is the system’s way of saying, “I’m working really hard.”

When GPU utilization peaks (95%), it produces heat rapidly.
Once temperature crosses a safe limit, thermal throttling activates, a safety mechanism that slows performance to protect the hardware.

This throttling explains why the AI response halted and fan speed decreased shortly after.

📽️ Watch the video below that explains this situation.
Hardware strain during model inference

💡 Why This Is Normal

"Small" in AI doesn't mean "light" for consumer hardware.
A quick comparison:
• My GPU (MX130) -> ~0.4 TFLOPS
• Modern AI GPU (RTX 4090) -> ~82 TFLOPS
A gap of nearly 200x.

So, even though my laptop is capable of running local inference, it has trouble maintaining it, particularly when the model needs to continue executing billions of operations per second for longer outputs.

🔚 Final Thought

Running AI locally isn't just about generating text. It teaches you how your CPU, GPU, RAM, and thermal system manage the workload of modern AI models in real time.

Now, every time my laptop fan spins up, I know it's just trying to think really hard. 😅

Did you ever run into a similar issue? If yes, how did you tackle it?

If you found this post useful or learned something new, drop a ❤️ and share your thoughts in the comments. I'd love to hear your experience.

Feel free to reach out on

💌 Email

💻 GitHub

localStorage - One of the simplest yet most powerful browser features every front-end developer should master

Richa — Wed, 12 Nov 2025 07:19:15 +0000

A Practical Guide to Using localStorage in JavaScript (With Mini Project) - DEV Community

When we reload a website, JavaScript variables disappear; that’s because the web is stateless by...

dev.to

A Practical Guide to Using localStorage in JavaScript (With Mini Project)

Richa — Wed, 12 Nov 2025 07:17:15 +0000

When we reload a website, JavaScript variables disappear; that’s because the web is stateless by default so the browser doesn’t remember anything once the page is refreshed.
But what if you want your site to remember user preferences, saved items, or a theme choice?

That’s where localStorage comes in.

💡 What Is `localStorage`?

It is part of the Web Storage API, which lives in the Window object, meaning it’s only available in browser JavaScript (not Node.js).

Key features:

Data persists even after closing the browser
Storage limit: around 5–10 MB (varies by browser)
Accessible only by JavaScript running on the same domain
Stores data as strings only

🎯 Why Use localStorage?

✔️ It’s great for small, persistent, non-sensitive data such as:

Theme or layout preferences
Remembering form inputs or drafts
Storing app state (like sidebar open/closed)
Caching API responses temporarily

❌ It’s not meant for:

Sensitive data (like tokens or passwords)
Large or complex datasets (use IndexedDB for that)

🛠️ Basic Operations

1. Store Data

localStorage.setItem("name", "Richa");

2. Retrieve Data

const name = localStorage.getItem("name");
console.log(name); // "Richa"

3. Remove a Key

localStorage.removeItem("name");

4. Clear All Data

localStorage.clear();

5. Retrieve the name of a key

let keyName = localStorage.key(0); // Gets the name of the key at index 0
console.log(keyName); // name

Each value is stored as a string. If you try to store an object directly, you’ll just get [object Object].

📥 Storing Objects and Arrays (The Right Way)

Since localStorage only supports strings, you must convert structured data into strings using JSON.stringify() and parse it back with JSON.parse().

const user = { name: "Richa", role: "Developer" };

// Save
localStorage.setItem("user", JSON.stringify(user));

// Retrieve
const data = JSON.parse(localStorage.getItem("user"));
console.log(data.role); // "Developer"

⚙️ Real-World Example: Persistent Theme Switcher

👇Check out mini project that describes the use case of localStorage
Persistent Theme Switcher

✅ Try switching themes, reload the page, and see the magic; your preference stays saved!

🆚 Comparison: `localStorage` vs `sessionStorage` vs `IndexedDB`

Feature	localStorage	sessionStorage	IndexedDB
Persistence	Until manually cleared	Until tab closes	Persistent
Data Type	Strings only	Strings only	Objects, Blobs, Files
Size Limit	~5–10 MB	~5 MB	Hundreds of MBs
Use Case	Preferences, small cache	Temporary state	Large structured data
Access Speed	Fast (sync)	Fast (sync)	Slower (async)

🔚 Conclusion

So that’s it for localStorage!
It may look simple, but it’s an incredibly useful tool when used wisely.
It gives your web app memory, a way to remember who your user is and what they prefer, even after they close the browser.

I hope you found this mini tutorial helpful.

If you enjoyed this post or learned something new, drop a ❤️ and follow me for more beginner-friendly guides on web development.

And if you have any questions or feedback, feel free to reach out on

💌 Email

💡 LinkedIn

💻 GitHub

A simple HTML feature that transformed messy JS into clean, reusable UI components.

Richa — Mon, 13 Oct 2025 08:00:53 +0000

The <template> Tag: A core HTML Feature That Simplified My JS - DEV Community

I’m currently developing a conversational AI web application. I’ve been building it completely from...

dev.to

The

Feature	PostgreSQL	MySQL	SQL Server	Oracle
Basic CTE	✅	✅ (v8.0+)	✅	✅
Recursive CTE	✅	✅ (v8.0+)	✅	✅
Materialized CTE	✅ (v12+)	❌	❌	❌

Forem: Richa

Explore a clear, real-world case of what happens when you run a local AI model. A classic example of hardware strain during model inference

When a “Small” AI Model Pushes Your Hardware to Its Limits - DEV Community

When a “Small” AI Model Pushes Your Hardware to Its Limits

🧠 What Was Really Happening

💡 Why This Is Normal

🔚 Final Thought

localStorage - One of the simplest yet most powerful browser features every front-end developer should master

A Practical Guide to Using localStorage in JavaScript (With Mini Project) - DEV Community

A Practical Guide to Using localStorage in JavaScript (With Mini Project)

💡 What Is localStorage?

🎯 Why Use localStorage?

🛠️ Basic Operations

📥 Storing Objects and Arrays (The Right Way)

⚙️ Real-World Example: Persistent Theme Switcher

🆚 Comparison: localStorage vs sessionStorage vs IndexedDB

🔚 Conclusion

A simple HTML feature that transformed messy JS into clean, reusable UI components.

The <template> Tag: A core HTML Feature That Simplified My JS - DEV Community

⛔ The Problem: Too Much HTML in JavaScript

✅ The Solution: <template> Tag

🛠️ How It Works

🧐 Why is this important?

🔚 Conclusion

A beginner-friendly guide on how to write better git commit messages with Conventional Commits

Stop Writing Bad Commit Messages: Use Conventional Commits Instead - DEV Community

Stop Writing Bad Commit Messages: Use Conventional Commits Instead

What is Conventional Commit?

Format & Syntax

Examples for Each Type

Imagine you’re building a Chat App.

Why It Matters

Conclusion

𝗜 𝘁𝗿𝗶𝗲𝗱 𝗢𝗹𝗹𝗮𝗺𝗮 𝘁𝗼 𝗿𝘂𝗻 𝗟𝗟𝗠𝘀 𝗹𝗼𝗰𝗮𝗹𝗹𝘆, 𝗮𝗻𝗱 𝗶𝘁'𝘀 𝗮𝗺𝗮𝘇𝗶𝗻𝗴!

Ollama: How to Easily Run LLMs Locally on Your Computer

Ollama: How to Easily Run LLMs Locally on Your Computer

🔍 What Is Ollama?

💻 Installing Ollama

🛠️ Exploring the CLI

⁉️ Why I use llama3.2:1b model?

💬 Final Thoughts

Why CTEs Make Complex SQL Queries Easy

Mastering CTEs in SQL: A Beginner's Guide with Real-World Use Cases (PostgreSQL Edition).

Mastering CTEs in SQL: A Beginner's Guide with Real-World Use Cases (PostgreSQL Edition).

🔍 What is a CTE?

💡 Real-World Use Cases of CTEs

1. Simplifying Complex Joins & Filters

2. Recursive Queries

3. Updating with Joins (a CTE hack)

📚 Rules, Restrictions & Tips

✅ Rules

⚠️ Restrictions & Performance Tips

🔍 RDBMS Support

🖥️ My SQL and PostgreSQL Learning Project

Conclusion

💡 What Is `localStorage`?

🆚 Comparison: `localStorage` vs `sessionStorage` vs `IndexedDB`

✅ The Solution: `<template>` Tag

⁉️ Why I use `llama3.2:1b` model?