open-thoughts/OpenThinker-Agent-v1

OpenThinker-Agent-v1 by OpenThoughts is an 8 billion parameter model post-trained from Qwen3-8B, specifically optimized for agentic tasks. It excels in environments like Terminal-Bench 2.0 and SWE-Bench, demonstrating strong performance in automated problem-solving and code-related challenges. The model was developed using a two-stage process involving supervised fine-tuning and reinforcement learning on curated datasets. It is designed for applications requiring autonomous task execution and robust agent capabilities.