Soup turns the pain of LLM fine-tuning into a simple workflow. One config, one command, done.

A fully-contained, ready-to-run environment to finetune Llama 3 model with custom dataset and run inference on the fine-tuned models

Prompt-Brickz: a Lego-style prompt-builder that lets you snap personas, goals, and styles together—complete with a Perspective-Shift-Effect double pass for instant peer-review polish.

Assigns a respective class name to an uploaded image in the .jpg or .jpeg extensions format.

Spam Email Classification using LoRA Fine-tuned Transformers: High-performance spam email classification using LoRA-adapted transformer models (ELECTRA, RoBERTa). Achieves 99.4%+ accuracy with parameter-efficient fine-tuning on 83K+ emails.

Development of guess a word game with NLP and optimal strategies search

using chat gpt to fine tune custom data model

From Dataset to Optimization: A Benchmarking Framework for Information Retrieval in the Particle Accelerator Domain

Build precise goal prompts for Claude CLI agents using a five-section template to ensure audit-friendly execution and efficient token usage.