unknown

🚀 Introducing MiroThinker-1.7 & MiroThinker-H1

Today, we release the latest generation of our research agent family: MiroThinker-1.7 and MiroThinker-H1.

Our goal is simple but ambitious: move beyond LLM chatbots to build heavy-duty, verifiable agents capable of solving real, critical tasks. Rather than merely scaling interaction turns, we focus on scaling effective interactions — improving both reasoning depth and step-level accuracy.

Key highlights:

🧠 Heavy-duty reasoning designed for long-horizon tasks
🔍 Verification-centric architecture with local and global verification
🌐 State-of-the-art performance on BrowseComp / BrowseComp-ZH / GAIA / Seal-0 research benchmarks
📊 Leading results across scientific and financial evaluation tasks

Explore MiroThinker:

Hugging Face: 
Github:

MiroMind AI announces the release of MiroThinker-1.7 and MiroThinker-H1, described as the latest generation of their research agent family.