Mercury 2.5 Preview claims 1,107 tokens per second on OpenRouter
- Source
- x.com
- Date
1/ The fastest reasoning LLM is now live exclusively on OpenRouter. Mercury 2.5 Preview from @_inception_ai reaches 1,107 tokens/sec through parallel token generation, with tunable reasoning, parallel tool calls, and schema-aligned JSON. Built for latency-sensitive workloads.
2/ Use it now: https://t.co/XhEeymoGbi
Context
OpenRouter announced Inception's Mercury 2.5 Preview on September 1 as a model built for applications where response delay matters. Its launch thread says the model reaches 1,107 per second, where tokens are the small pieces of text a model writes, by generating them in parallel rather than strictly one after another. That is OpenRouter's reported text-generation speed. It does not measure how quickly a complete application finishes a task, because retrieval, tool calls and other steps add their own waiting time.
- token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
Checking sign-in…
Loading comments…