Inception introduces Mercury 3 with claimed speed above 1,000 tokens per second

First reported by RuntimeWire at · Updated · 2 sources

Inception says its diffusion model delivers benchmark gains over rival models. According to Unite.AI, the company claims Mercury 3 beats GPT-6 Luna and Claude Haiku 5.5 across multiple benchmarks while running four to eight times faster. Unite.AI reports launch prices of $0.10 per million input tokens and $0.35 per million output tokens.

  • RuntimeWire reports a claimed context window of 1 million tokens.
  • Unite.AI dates the launch to October 19, 2026.

Covered by 2 publishers within 1 hour of the first report.

Reporting2

RuntimeWire Inception launches Mercury 3 at 1,068 tokens per second · Ryan Merket
Unite.AI Inception Releases Mercury 3, Its Flagship Diffusion LLM · Jonas Reeve, Cognitive AI & AGI, AI Research Agent

Related

Replit makes OpenAI its default over Anthropic, CEO says
FTC subpoenas OpenAI and Anthropic over AI safety incidents
Andrej Karpathy says only 5,000 people have seen what AI can really do
OpenAI and Anthropic poach Trump officials as industry fights for Washington’s trust

Topics OpenAIAnthropic