Fine-tuning a 1.7B model at 3.2 GB VRAM — building FineTune Studio
PRANJUL RATHOURDev.to (EN Zone)
1 views
FineTune Studio exists because every fine-tuning tutorial I found assumed a rented A100 and a notebook full of half-explained flags. I wanted something a student could run: upload a dataset, validate it, launch a real QLoRA job, watch the loss stream live, then see — not assume — whether the fine-tuned model actually improved.
The numbers that matter
17 backend endpoints, 107 / 107 tests passing.
Base model Qwen3-1.7B, peak VRAM ~3.2 GB.
3 inference paths: local, vLLM, or a Hugging Face Space — pick per deployment.
7 frontend pages: Dashboard, Datasets, Training, Live Logs, Evaluation, Experiments, Settings.
Dataset validation before a single GPU cycle
Most fine-tuning failures I've seen aren't hyperparameter problems, they're dataset problems — malformed JSON, duplicate examples, a prompt template mismatch. FineTune Studio validates the dataset on upload and reports the actual issue, not a stack trace three steps into training.
Live telemetry, not a log file to tail
Training streams loss, learning rate and throughput live over the same connection the frontend already holds — so a run's progress is a graph, not a terminal you keep alt-tabbing to.
The comparison is the whole point
If you can't show the before and after, you didn't fine-tune — you spent GPU hours. The Evaluation page runs the same prompts against the base model and the tuned model side by side. That comparison, more than the loss curve, is what tells you whether the run was worth it.
What I'd tell a student starting this
Don't buy a GPU first. Learn on the free tier and on small models — the memory limit teaches you more about your dependencies than any tutorial does. A 3B model with your own dataset and an honest eval beats a huge model behind an API you only rent for a demo.
Code and the full architecture: github.com/Pranjulrathour/FINETUNESTUDIO.
Pranjul Rathour · GenAI engineer, 3x hackathon winner, campus mentor. Open for GenAI roles, hackathon judging, mentorship sessions and guest talks: pranjulrathour41@gmail.com · Invite me to your campus
Portfolio & blog · LinkedIn · Bluesky · GitHub · Dev.to
Over the past few days, I was trying out gov.uk prototype system for my static website. Like every night owl, the unbranded template's white background was hurting my eyes. I googled for some sort of dark mode, found The National Archives Design System. It suits my purpose, has the dark mode, and i
WebForms.php 2.1 has been released as the PHP back-end implementation of WebForms Core 2.1.
This release is different from a typical porting story.
The PHP implementation was converted from the C# implementation of WebForms Core using DeepSeek, and then independently evaluated with Qwen.
The proc
When working with JavaScript functions, you will often hear two terms: parameters and arguments.
They are closely related, but they have different meanings.
What is a Parameter?
A parameter is a variable that we define inside the function's parentheses when creating a function.
It acts