Mercury 2.5 runs at 1,100 tokens/sec

Melvin Vivas · X post · 2026-09-02 · Open on X

Topics: Industry Trends & Job Market, LLM Fundamentals · Level: beginner

Summary

The creator reacts to the Mercury 2.5 release. Its team says it beats Mercury 2 on quality, especially on agentic tasks, and is faster at about 1,100 tokens/sec.

Key points

• Mercury 2.5 claims about 1,100 tokens/sec
• Claimed to beat Mercury 2 on both quality and speed ('pareto-dominated')
• Biggest quality gains claimed on agentic tasks

Resources mentioned

• Mercury 2.5 · tool · inceptionlabs.ai · paid
  A very fast language model, reported at about 1,100 tokens/sec with better agentic performance than Mercury 2.
  Also in: Jev + WebMCP Solves 100% of Benchmark Tasks at About 112× Lower Cost (Melvin Vivas on X · notes)