Pistis comes in 27B and 9B multimodal models built on Qwen3.6 and Qwen3.5 respectively, starting from large-scale multimodal supervised fine-tuning.
IDRL alternates on-policy distillationTraining a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.Full definition → and reinforcement learning inside one training loop, instead of running either alone or combining them in a static joint loss; the authors credit it with steadier optimization and better credit assignment on long agentic trajectories.
Each size ships as Pistis-Thinking, for deep multimodal reasoning, and Pistis-Agentic, which adds agentic trajectory data for planning and tool useA model's ability to call external functions — run code, search the web, edit files — instead of only generating text.Full definition → and is reported to be particularly strong at multimodal search.
Pistis-Auto-Harnessing iteratively optimizes the AI agentAn 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.Full definition →'s inference agent harnessThe scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.Full definition → and, per the authors, improves performance without updating model weights or increasing the interaction budget.
Terms in this piece · Glossary
distillation — Training a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.
tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
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.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
Why it matters
IDRL's alternating distillation/RL loop targets a real post-training pain point, unstable credit assignment on long agentic trajectories, and could inform how teams design their own post-training pipelines.
articleFrom PyTorch to the NPU: LLM-Agent-Driven Model Conversion Across Heterogeneous Inference RuntimesJianhao Su, Zhanwei Wu, Chia-Heng Tu, ShengTing Huang
articlePretraining and adapting a language model on a dependency-free stack: GPT-2 124M from random weights, reproduced against llm.c, and a clinical adapter for Qwen3-0.6BThang Tran (CloudKites AI Lab, New South Wales, Australia), Lan Dang (Monash Business School, Monash University, Victoria, Australia)
postInside GLM-5V-Turbo: dual-teacher vision distillation and 30-task RLZ.ai