Trained Tiny Tales GPT(30M model)from scratch and deployed in $15
For the last few weeks, I have been working on training an LLM from scratch and deploying it in production on Google Cloud Platform. Finally, I trained a 30 million parameter model on 1 billion tokens and deployed it as a web service. You can access the LLM using this site - https://kunalmishra.info The following steps were taken to build Tiny Tales GPT 1. Downloaded and preprocessed 8GB of dataset using multiprocessing library. 2. Tokenized the data using byte pair encoding to create 1 billion tokens sharded in different bin files. 3. Defined a training setup and trained the model…
In plain words
Trained Tiny Tales GPT is a 30-million parameter language model trained from scratch on 1 billion tokens and deployed as a web service. The model uses a LLaMA-style architecture and was trained on preprocessed text data using distributed training across GPUs. It is designed for users who want to interact with a smaller, production-deployed language model. The project demonstrates the complete pipeline from dataset preparation and tokenization through model training and cloud deployment.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
For the last few weeks, I have been working on training an LLM from scratch and deploying it in production on Google Cloud Platform. Finally, I trained a 30 million parameter model on 1 billion tokens and deployed it as a web service. You can access the LLM using this site - https://kunalmishra.info The following steps were taken to build Tiny Tales GPT 1. Downloaded and preprocessed 8GB of dataset using multiprocessing library. 2. Tokenized the data using byte pair encoding to create 1 billion tokens sharded in different bin files. 3. Defined a training setup and trained the model on a small version of the LLaMA model architecture with 30 million parameters. 4. The training was done using Distributed Data-Parallel on two A-100 GPUs provided by JarvisLabs.ai (they are most cost-optimized) 5. After the training is done, an inference script is created to predict the tokens from the trained model given the input context vector. 6. Developed REST-based API service using Flask framework to interact with the inference service to the end user. 7. Finally used GCP's virtual machines, instance groups, load balancers, and DNS services to deploy the service on the internet.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, April 2024
the whole month →
Supabase▲2,328The Postgres developer platform is now generally available
Dev tools · 2024 · supabase.com
Cal.com Platform▲1,602Build your pixel-perfect booking experience with Atoms
Dev tools · 2024 · cal.com
deco.cx 2.0▲1,080Build web apps 10x faster with Deno, JSX, TS & Tailwind
Dev tools · 2024 · decocms.com
