Arcee AI: Virtuoso Medium V2

Arcee AI's Virtuoso-Medium-v2, a 32B model distilled from DeepSeek-v3 and merged onto a Qwen 2.5 backbone, demonstrates a strong balance of capability and deployability. Created on May 5, 2025, it consistently ranks among the fastest models, achieving the 91st percentile across six benchmarks, and offers competitive pricing, placing in the 55th percentile. Its reliability is exceptional, boasting a perfect 100th percentile, indicating minimal technical failures and consistent evaluable responses. Across benchmarks, Virtuoso-Medium-v2 shows particular strength in Ethics, achieving perfect 100% accuracy, making it the most accurate model at its price point and among models of similar speed. It also excels in General Knowledge with 99% accuracy and strong Instruction Following at 61% accuracy (70th percentile). While its Email Classification accuracy is high at 97%, its percentile ranking (50th) suggests this is a common performance level for models in this category. Its Coding (83%) and Reasoning (68%) performances are solid, placing it in the 62nd and 64th percentiles respectively. The model's aggressive quantization options and 128k context length make it suitable for enterprise chat, technical writing, and medium-complexity code drafting, offering a sharp, factual successor to its predecessor.