Hugging Face's logo
Pankaj Pandey's picture
šŸ—ļø Building on HF

Pankaj Pandey

pankajpandey-dev
5 3 20
kaushik-harsh-99's profile picture or99's profile picture keke-420's profile picture
Ā·

AI & ML interests

Natural Language Processing, Text Generation, Large Language Models, Quantization, Fine-Tuning, RLHF, Model Merging.

Recent Activity

repliedto their post about 1 month ago
šŸ‡®šŸ‡³ Qwen3.5-9B Hindi Instruct ā€” it stops thinking in English Ask base Qwen3.5-9B a question in Hindi and it burns hundreds of tokens thinking in English inside its think block before a single Devanagari word appears ā€” then code-switches in the answer. I fine-tuned it to close the think block instantly and reply in pure, native Hindi. āœ… Model (16-bit): https://huggingface.co/pankajpandey-dev/qwen3.5-9b-hindi-instruct āœ… GGUF (Q4/Q5/Q8): https://huggingface.co/pankajpandey-dev/qwen3.5-9b-hindi-instruct-GGUF āœ… Try it in the browser: https://huggingface.co/spaces/pankajpandey-dev/qwen3.5-9b-hindi-demo Recipe: Unsloth + LoRA (r=16, response-only loss) on 12.9k Hindi pairs ā€” AI4Bharat anudesh + dolly-hi + wikiHow-hi + Aya Hindi (human-written). The Q4_K_M is 5.4 GB and runs on a plain laptop CPU. New in this run vs my earlier models: mixed in long-form native sources (wikiHow) after my last eval showed the fine-tune traded detail for conciseness ā€” this one keeps answers detailed and native. Part of my weekly šŸ‡®šŸ‡³ Hindi LLM Series. Feedback welcome šŸ™ #Hindi #IndicNLP #Qwen #GGUF #LocalLLM #Unsloth
repliedto their post about 1 month ago
šŸ‡®šŸ‡³ Qwen3.5-9B Hindi Instruct ā€” it stops thinking in English Ask base Qwen3.5-9B a question in Hindi and it burns hundreds of tokens thinking in English inside its think block before a single Devanagari word appears ā€” then code-switches in the answer. I fine-tuned it to close the think block instantly and reply in pure, native Hindi. āœ… Model (16-bit): https://huggingface.co/pankajpandey-dev/qwen3.5-9b-hindi-instruct āœ… GGUF (Q4/Q5/Q8): https://huggingface.co/pankajpandey-dev/qwen3.5-9b-hindi-instruct-GGUF āœ… Try it in the browser: https://huggingface.co/spaces/pankajpandey-dev/qwen3.5-9b-hindi-demo Recipe: Unsloth + LoRA (r=16, response-only loss) on 12.9k Hindi pairs ā€” AI4Bharat anudesh + dolly-hi + wikiHow-hi + Aya Hindi (human-written). The Q4_K_M is 5.4 GB and runs on a plain laptop CPU. New in this run vs my earlier models: mixed in long-form native sources (wikiHow) after my last eval showed the fine-tune traded detail for conciseness ā€” this one keeps answers detailed and native. Part of my weekly šŸ‡®šŸ‡³ Hindi LLM Series. Feedback welcome šŸ™ #Hindi #IndicNLP #Qwen #GGUF #LocalLLM #Unsloth
updated a model about 1 month ago
pankajpandey-dev/qwen3.5-9b-hindi-instruct
View all activity

Organizations

scikit-learn's profile picture LocalLLaMA's profile picture Hugging Science's profile picture nanochat students's profile picture