GGUF export of the Hinglish-Gemma-4B-E4b LoRA adapter, for use with llama.cpp and Ollama. It makes Gemma 4 E4B chat like your desi best friend — in casual Hinglish.
This is the adapter only (73 MB) — pair it with any Gemma 4 E4B base GGUF, e.g. unsloth/gemma-4-E4B-it-GGUF (Q4_K_M recommended).
llama.cpp
bash
1llama-cli -m gemma-4-E4B-it-Q4_K_M.gguf \2 --lora Hinglish-Gemma-4B-E4b-lora-f16.gguf \3 -sys "You are Yaar, a friendly desi buddy. You always reply in casual Hinglish (Hindi written in English letters, mixed with English), like a close friend texting. Keep replies short, warm and natural."\4 -p "yaar kal exam hai aur kuch padha nahi"
# Modelfile
FROM ./gemma-4-E4B-it-Q4_K_M.gguf
ADAPTER ./Hinglish-Gemma-4B-E4b-lora-f16.gguf
SYSTEM "You are Yaar, a friendly desi buddy. You always reply in casual Hinglish (Hindi written in English letters, mixed with English), like a close friend texting. Keep replies short, warm and natural."
ollama create yaar -f Modelfile && ollama run yaar "yaar aaj mood off hai"
🧠 Model Lineage
Fine-tuned by Vedant Rajpurohit from Google's Gemma 4 E4B on the Abhishekcr448/Hinglish-Everyday-Conversations-1M dataset (~52k curated examples, QLoRA r=16). Full training details on the adapter card.
If you use this, a mention or backlink is appreciated 💛
