arcee-ai/Arcee-Nova

Arcee-Nova is a 72.7 billion parameter instruction-tuned language model developed by arcee-ai, based on a merge of Qwen2-72B-Instruct with a custom generalist dataset. It features a 131,072 token context length and is optimized for a wide range of tasks including reasoning, creative writing, and coding. Evaluated on the OpenLLM Leaderboard 2.0 stack, it demonstrates performance approaching that of GPT-4 (May 2023).