OpenBuddy/openbuddy-nemotron-70b-v23.1-131k

Warm
Public
70B
FP8
32768
License: llama3.1
Hugging Face
Overview

OpenBuddy/openbuddy-nemotron-70b-v23.1-131k: Multilingual Chatbot

This model is a 70 billion parameter instruction-tuned chatbot developed by the OpenBuddy team, based on the powerful nvidia/Llama-3.1-Nemotron-70B-Instruct architecture. It is specifically designed for conversational AI in both English and Chinese, making it a versatile tool for multilingual applications.

Key Capabilities & Features

  • Multilingual Support: Optimized for chat interactions in English and Chinese.
  • Instruction-Tuned: Follows a specific prompt format (<|role|>system<|says|>...<|end|>) for structured and effective conversations, compatible with transformers and vllm fast tokenizers.
  • Ethical Guidelines: Programmed to provide helpful, respectful, and honest responses, avoiding harmful, political, unethical, or biased content.
  • Knowledge Cutoff: Possesses vast knowledge up to April 2023, operating without internet access.

Performance Highlights

Evaluations on the Open LLM Leaderboard show an average score of 39.08. Specific metrics include:

  • IFEval (0-Shot): 75.55
  • BBH (3-Shot): 53.19
  • MMLU-PRO (5-shot): 46.38

Intended Use Cases

This model is suitable for developers building chatbots or conversational agents that require strong performance in both English and Chinese. Its instruction-following capabilities make it ideal for applications where structured and safe responses are critical. Users should be aware of the inherent limitations and disclaimers associated with all large language models, particularly regarding critical or high-stakes scenarios.