moushi21/agent-bench-merged12
moushi21/agent-bench-merged12 is a 4 billion parameter Qwen3-based model, created by moushi21, specifically optimized for agentic tasks. This model was developed using the TIES-Merging method to combine specialized LoRA adapters for ALFWorld and DBBench (SQL) tasks. It demonstrates balanced performance on agentic benchmarks, achieving 0.60 Pass@1 on ALFWorld and 0.5353 on DBBench. The model is designed for direct inference in agent-based applications requiring strong reasoning in interactive environments and database interactions.
Loading preview...
Overview
moushi21/agent-bench-merged12 is a 4 billion parameter model built upon the Qwen/Qwen3-4B-Instruct-2507 base, specifically engineered for enhanced performance in agentic tasks. This model was created by merging specialized LoRA adapters using the TIES-Merging method via Mergekit, integrating expertise from ALFWorld trajectories and DBBench (SQL) tasks.
Key Capabilities
- Agentic Task Optimization: Fine-tuned to excel in complex agent environments, particularly those involving interactive decision-making and database querying.
- Specialized Merging: Utilizes the TIES-Merging method to combine distinct LoRA adapters, ensuring a balanced integration of specialized skills.
- Direct Inference: Provided as full model weights, eliminating the need to load separate adapters for deployment.
Performance Highlights
The model exhibits strong performance across its target agentic benchmarks:
- ALFWorld: Achieves a Pass@1 score of 0.60.
- DBBench: Scores 0.5353.
Training Data & Licensing
The source models were fine-tuned on specific datasets:
- ALFWorld:
u-10bei/sft_alfworld_trajectory_dataset(v1 to v5) - DBBench:
u-10bei/dbbench_sft_dataset_react(v1 to v4)
Both datasets are distributed under the MIT License. Users must comply with the MIT license and the base model's original terms of use.