Quantized GGUF builds of pankajpandey-dev/gemma-3-1b-hindi-instruct for CPU / edge inference. Quants: Q4_K_M (~806MB), Q5_K_M (~851MB), Q8_0 (~1.07GB).

Files carry the base name gemma-3-1b-it.* but contain these fine-tuned weights.

Ollama

ollama run hf.co/pankajpandey-dev/gemma-3-1b-hindi-instruct-GGUF:Q4_K_M

llama.cpp

llama-cli -m gemma-3-1b-it.Q4_K_M.gguf -p "एक छोटे बच्चे को गुरुत्वाकर्षण सरल हिंदी में समझाइए।" --temp 0.4

Part of my 🇮🇳 Hindi LLM Series — weekly experiments adapting small models to Indian languages.

Available formats

Repo	Format	Use
`...-hindi-instruct`	Merged 16-bit	Transformers
`...-hindi-instruct-GGUF`	Q4_K_M / Q5_K_M / Q8_0	Ollama, llama.cpp, CPU
`...-hindi-instruct-lora`	LoRA adapter	Method artifact

Training

Base: google/gemma-3-1b-it (text-only path)
Method: LoRA (r=32, α=32, all attn+MLP projections), response-only loss
Data: AI4Bharat indic-instruct-data-v0.1 — anudesh + dolly (Hindi), chrF≥50 filtered, balanced 6k
Schedule: 2 epochs, LR 2e-4, effective batch 8 · single T4 (Kaggle, free), fp32, ~167 min · Unsloth + TRL

Recommended decoding: temperature=0.4, top_p=0.9, repetition_penalty=1.3.

Evaluation

प्रश्न: एक छोटे बच्चे को गुरुत्वाकर्षण सरल हिंदी में समझाइए। उत्तर: PASTE_YOUR_BEST_CLEAN_OUTPUT_HERE

Limitations

A 1B model — Hindi fluency is solid; coherence/factual reliability are bounded by scale. Best for short instructions, simple Q&A, and edge/demo use. A Gemma-3-4B Hindi version is the planned next step.

pankajpandey-dev/gemma-3-1b-hindi-instruct-GGUF

🇮🇳 Gemma-3-1B Hindi Instruct — GGUF