The landscape of open speech recognition models has become more diverse in 2026, with several models now offering similar performance. Models such as Cohere Transcribe, IBM Granite Speech 4.1, ARK-ASR, and MOSS-Transcribe have narrowed the gap in word error rate, according to the Hugging Face Open ASR Leaderboard. This shift has made it more difficult to determine a clear ranking among the top models. The models are also being compared on factors such as language coverage, streaming latency, and licensing terms. This increased competition may lead to improved performance and more choices for developers. This development matters as it could lead to more effective and accessible speech recognition technology in various applications.