Dhee-NxtGen-Qwen3-Tamil-v2

Model Description

Dhee-NxtGen-Qwen3-Tamil-v2 is a large language model designed for natural Tamil language understanding and generation.
Built on the Qwen3 architecture, it has been fine-tuned for assistant-style reasoning, multi-turn dialogue, and task-oriented responses.

This model is part of DheeYantra’s multilingual LLM series, developed in collaboration with NxtGen Cloud Technologies Pvt. Ltd., to power Indic conversational AI applications.

Key Features

Context-aware and fluent Tamil text generation
Trained for assistant-like conversations and reasoning
Suitable for chatbots, summarization, and story/narrative generation
Fully compatible with 🤗 Hugging Face Transformers
Optimized for VLLM for ultra-fast inference

Example Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "dheeyantra/dhee-nxtgen-qwen3-tamil-v2"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, trust_remote_code=True)

# Example prompt
prompt = """<|im_start|>system
You are a helpful assistant.<|im_end|>
<|im_start|>user
நீங்கள் எனக்காக ஒரு நேர்முகத்தை ஒதுக்க முடியுமா?<|im_end|>
<|im_start|>assistant
"""

inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=150)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Intended Uses & Limitations

Intended Uses

Tamil conversational chatbots and assistants
Function-calling and structured response generation
Story generation and summarization in Tamil
Natural dialogue systems for Indic AI applications

Limitations

May generate inaccurate or biased responses in rare cases
Performance can vary on out-of-domain or code-mixed inputs
Primarily optimized for Tamil; other languages may produce less fluent results

VLLM / High-Performance Serving Requirements

For high-throughput serving with vLLM, ensure the following environment:

GPU with compute capability ≥ 8.0 (e.g., NVIDIA A100)
PyTorch 2.1+ and CUDA toolkit installed
For V100 GPUs (sm70), vLLM GPU inference is not supported; CPU fallback is possible but slower.

Install dependencies:

pip install torch transformers vllm sentencepiece

Run vLLM server:

vllm serve   --model dheeyantra/dhee-nxtgen-qwen3-tamil-v2   --host 0.0.0.0   --port 8000

License

Released under the Apache 2.0 License.

Developed by DheeYantra in collaboration with NxtGen Cloud Technologies Pvt. Ltd.

dheeyantra/dhee-nxtgen-qwen3-tamil-v2