Intel / post

Ling-2.6-1T Lands on OpenRouter With a Fast-Thinking Cost Cut

Source: Ant Ling
Date: Published May 7, 2026

Context

Ant Ling announces that its trillion-parameter Ling-2.6-1T instruct model, built by InclusionAI, is available through OpenRouter. The company describes the model using a 'fast thinking' approach that generates short reasoning rather than long chains of thought, and says this cuts cost by about 75% while maintaining what it calls state-of-the-art performance on the AIME 26 math competition and , a benchmark that scores whether a model can resolve real GitHub issues. The specific benchmark scores are not stated in this thread, so the state-of-the-art claim reflects the company's own characterization rather than a number that can be checked here.

Terms in this piece · Glossary

inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
SWE-bench — The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Read the source x.com