Dhee-NxtGen-Qwen3-Maithili-v2

Model Description

Dhee-NxtGen-Qwen3-Maithili-v2 is a large language model designed for natural Maithili language understanding and generation.
It is based on the Qwen3 architecture and fine-tuned for assistant-style, function-calling, and reasoning-based conversational tasks.

Developed by DheeYantra in collaboration with NxtGen Cloud Technologies Pvt. Ltd., this model powers intelligent Maithili conversational systems and reasoning agents.

Key Features

Fluent and context-aware Maithili text generation
Optimized for assistant and reasoning conversations
Supports open-ended generation, summarization, and dialogue
Fully compatible with 🤗 Hugging Face Transformers
Optimized for VLLM for high-performance inference

Example Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "dheeyantra/dhee-nxtgen-qwen3-maithili-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
की अहाँ हमरा एकटा अपॉइन्टमेंट सिड्यूल करी देब?<|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

Maithili conversational chatbots and assistants
Function-calling and structured response generation
Story generation and summarization in Maithili
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 Maithili; 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-maithili-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-maithili-v2