Mercury 2 Model Released, Focused on High-Speed Inference and Cost-Performance Ratio

robot
Abstract generation in progress

AIMPACT News, April 28 (UTC+8), Stefano Ermon tweeted on April 28, 2026, announcing that the Mercury 2 model is ready. The model is GPU-deployed, positioned at the same speed level as solutions like Cerebras, with the capacity to handle production traffic. The tweet targets users abandoned by Cerebras, claiming Mercury 2 offers inference speeds of over 1,000 tokens/sec, Haiku/Flash-level quality, priced at $0.75 per million output tokens. Contact email: sid@inceptionlabs.ai. (Source: InFoQ)

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin