Fine-tuning a 1.7B model at 3.2 GB VRAM — building FineTune Studio

Pranjul Rathour
Pranjul Rathour

Originally published at https://pranjulrathour.scult.in/blog/finetune-studio-qlora-on-a-budget. That copy is the canonical version and gets updates first.

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.

Presenting to a room
Presenting to a room

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.

Pranjul Rathour
Pranjul Rathour

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.

From my carousels
1,211 Prompts, Every One Stamped
1,211 Prompts, Every One Stamped, slide 11,211 Prompts, Every One Stamped, slide 2
1,211 Prompts, Every One Stamped, slide 31,211 Prompts, Every One Stamped, slide 4
Full carousel on Instagram and LinkedIn.
Pranjul Rathour
Pranjul Rathour
GenAI engineer, Kanpur · 3x first-prize hackathon winner · campus mentor
I ship production RAG pipelines, fine-tune LLMs and build agentic AI products end to end. I lead engineering at SCULT INDIA for a 14-member team and have mentored 200+ students through TechVerse Enclave.
Open to: GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges.
On stage, at hackathons and on campus
Presenting KrishGyan, farming advice in your voice and language
Presenting KrishGyan, farming advice in your voice and language
At an Integral Startup Foundation hackathon
At an Integral Startup Foundation hackathon
Pitching at CSJMU, Kanpur
Pitching at CSJMU, Kanpur

Comments

Post a Comment

Popular posts from this blog

Forming a hackathon team: roles, skills and the mistake most teams make

Image
At an Integral Startup Foundation hackathon Originally published at https://pranjulrathour.scult.in/blog/hackathon-team-roles-and-formation . That copy is the canonical version and gets updates first. Team TechVerse — Raman Shukla, Saksham Gupta and me — won three first prizes in 2025 not because we were the strongest coders in the room, but because we had stopped overlapping. When I mentor teams before a hackathon, formation is the first thing I fix. In a packed college auditorium The four roles Builder — owns the core feature and the demo path. Usually backend or the AI piece. Interface — owns what the judge sees: the front end, the flow, the seeded data. Storyteller — owns the problem statement, the pitch, the slides and the Q&A plan; also does user research in hour one. Integrator — keeps the pieces working together, manages the repo, runs the demo rehearsals. In a team of three, the builder or interface person takes this. The mistake Five friends who all write R...
Read more

Hello from Kanpur: what I build, and what I'll write about here

Image
On the mic Originally published at https://dev.to/pranjulrathour/hello-from-kanpur-what-i-build-and-what-ill-write-about-here-f2j . That copy is the canonical version and gets updates first. I'm Pranjul Rathour, a GenAI engineer from Kanpur, India. This is the first post on this blog, so here is who is writing, what I have actually shipped, and what you can expect if you follow along. What I build Between February and July 2026, at Next Upgrad Web Solutions, I designed, built and shipped five production AI applications end to end, alone, and deployed all of them with Docker, CI and observability: RAG.NextUpgrad : a retrieval-augmented generation service with hybrid dense plus BM25 retrieval, cross-encoder reranking, streaming cited answers, and a confidence gate that refuses to call the model when the evidence is weak. It runs in about 220 MB of memory, inside a free 512 MB instance. FineTune Studio : fine-tune open models with QLoRA in a browser. Upload a dataset, validate ...
Read more

I built 15 free tools, 1,211 prompts and a 50,000-skill library — here's what's inside tools.scult.in

Image
Carousel · 12 slides I built every part of tools.scult.in myself — 15 browser tools, 1,211 version-stamped prompts, a 50,000+ agent-skills library, a public MCP server, and 40 retro photo prompts where I'm the face in every example. All of it free. No signup, no ads, no upload step. Swipe through for what's inside and why I built it this way. Link in the last slide. Slide 1 of 12 Slide 2 of 12 Slide 3 of 12 Slide 4 of 12 Slide 5 of 12 Slide 6 of 12 Slide 7 of 12 Slide 8 of 12 Slide 9 of 12 Slide 10 of 12 Slide 11 of 12 Slide 12 of 12 Pranjul Rathour GenAI engineer, Kanpur · 3x first-prize hackathon winner · campus mentor I ship production RAG pipelines, fine-tune LLMs and build agentic AI products end to end. I lead engineering at SCULT INDIA for a 14-member team and have mentored 200+ students through TechVerse Enclave. Open to: GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges. Email: pranjulrathour41@gmail.com Portfolio & ...
Read more