Axion 1.0: Open Agent Foundation Models, Simplified
Paper · https://arxiv.org/abs/2611.03233
We introduce Axion 1.0, a fully open agent foundation model that combines strong and broad agentic capabilities with a remarkably simple training recipe. Axion 1.0 is trained in only three stages: two lightweight supervised fine-tuning stages followed by reinforcement learning, without dedicated mid-training, direct preference optimization (DPO), or on-policy distillation (OPD). The first SFT stage develops strong self-contained reasoning through model-aware data construction, where we synthesize challenging problems and adversarially strengthen them based on the base model's own successful solution strategies. The second stage extends these capabilities to tool-dependent reasoning through a scalable pipeline for constructing diverse and automatically verifiable tasks grounded in real-world data sources and deterministic computation. Finally, reinforcement learning further strengthens long-horizon reasoning and interaction with external tools and environments. Despite this compact training recipe, Axion 1.0 achieves strong performance across a broad suite of agentic benchmarks and is competitive with leading commercial models on several of them. Equally importantly, Axion 1.0 is open beyond model weights: we release the complete recipe, including data construction, generation, filtering, verification, prompts, and training code, enabling the entire development process to be reproduced and extended. Axion 1.0 demonstrates that strong agent foundation models can be both open and simple, offering a practical and reproducible path toward advancing open agentic intelligence.
First seen by Oriel on .
AI summary
Axion 1.0 is a fully open agent foundation model whose three-stage training recipe delivers broad agentic capabilities without mid-training, DPO, or OPD.
- The model uses two lightweight supervised fine-tuning stages followed by reinforcement learning.
- The first SFT stage builds self-contained reasoning with synthetic challenges adversarially strengthened from the base model’s successful strategies.
- The second SFT stage teaches tool-dependent reasoning through diverse, automatically verifiable tasks based on real-world data and deterministic computation.
- Reinforcement learning improves long-horizon reasoning and interaction with external tools and environments.
- The authors release weights, data construction and filtering methods, prompts, verification procedures, and training code for reproduction and extension.
#agent models #reinforcement learning #supervised fine-tuning #open models
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Points
Score reported by the source.
- Hugging Face Papers: 48, latest observation ; +10 over observed period since
Comments
Comment count reported by the source.
- Hugging Face Papers: 3, latest observation ; +1 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| Hugging Face Papers | 2611.03233 | https://arxiv.org/abs/2611.03233 Read discussion |