Pistis Technical Report

The Pistis model family, comprising 27B- and 9B-parameter multimodal large language models built on Qwen3.6 and Qwen3.5, was introduced in arXiv paper 2609.28554v1 using a post-training framework that combines large-scale multimodal supervised fine-tuning with Interleaved Distillation and Reinforcement Learning (IDRL). IDRL alternates on-policy distillation and reinforcement learning in a single training loop, yielding Pistis-Thinking for deep multimodal reasoning and Pistis-Agentic for long-horizon planning, iterative reasoning and tool use, with both scales outperforming their base models. The authors also present Pistis-Auto-Harnessing (PAH), which improves agent inference-harness performance without updating model parameters or increasing the interaction budget.

arXiv:2609.28554v1 Announce Type: new Abstract: We introduce the Pistis model family, comprising 27B- and 9B-parameter multimodal large language models built on Qwen3.6 and Qwen3.5, respectively, and developed through a general and scalable post-training framework. The framework first establishes a strong foundation through large-scale multimodal supervised fine-tuning SFT . Building on this SFT foundation, we propose Interleaved Distillation and Reinforcement Learning IDRL , a novel post-training paradigm that tightly integrates on-policy distillation and reinforcement learning within a single training loop. By alternating between the two objectives, rather than optimizing either in isolation or combining them in a static joint loss, IDRL enables more effective knowledge transfer, greater optimization stability, and more precise credit assignment for long-horizon agentic trajectories, leading to stronger performance while mitigating common capability trade-offs. At both model scales, the framework produces two specialized variants: Pistis-Thinking, designed to strengthen deep multimodal reasoning, and Pistis-Agentic, which additionally incorporates agentic trajectory data to support long-horizon planning, iterative reasoning, and tool use. Pistis-Agentic is particularly strong in multimodal search. Both scales outperform their corresponding base models. Beyond model-parameter optimization, we further introduce Pistis-Auto-Harnessing PAH , a system-level method that automatically improves the agent's inference harness through iterative optimization. Experiments demonstrate that PAH enhances the model performance without updating the model parameters or increasing the interaction budget.