unsloth/Qwen-AgentWorld-35B-A3B
Qwen-AgentWorld-35B-A3B is a 35.1 billion parameter language world model developed by Qwen, designed for agentic environment simulation. It is the first model to cover seven agent interaction domains, including tool calling, search, and software engineering, by predicting environment states through long chain-of-thought reasoning. This model excels as a generalizable and scalable simulator, trained natively for environment modeling rather than as a post-hoc adaptation.
Loading preview...
Qwen-AgentWorld-35B-A3B: A Native Language World Model
Qwen-AgentWorld-35B-A3B is a 35.1 billion parameter causal language model developed by Qwen, specifically engineered as a native world model for simulating agentic environments. Unlike general-purpose LLMs, its training objective from the Continual Pre-Training (CPT) stage onward focuses on environment modeling, followed by Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) to enhance simulation fidelity.
Key Capabilities & Features
- Seven Unified Domains: Covers a broad spectrum of agent interactions, including tool calling (MCP), Search, Terminal, Software Engineering (SWE), Android, Web, and OS environments, encompassing both text and GUI interactions.
- Environment Simulation: Predicts the next environment state based on an agent's action and interaction history, utilizing long chain-of-thought reasoning.
- Generalizable & Scalable: Demonstrates zero-shot generalization to out-of-domain environments and supports controllable perturbations for fictional-world construction.
- Agent Foundation Model: Provides an LWM RL warm-up that transfers to multi-turn, tool-calling agentic tasks across various benchmarks.
- Extended Context: Features a substantial context length of 262,144 tokens, crucial for multi-turn environment simulation.
Performance Highlights
On the AgentWorldBench, Qwen-AgentWorld-35B-A3B achieves an overall score of 56.39, performing competitively across domains like MCP (64.79) and SWE (65.63), showcasing its robust simulation capabilities compared to other large models.
Recommended Use Cases
This model is ideal for developers building or researching AI agents that require realistic and controllable environment simulations across diverse interaction domains. Its native world model design makes it particularly effective for tasks demanding accurate prediction of environmental responses to agent actions.