This model has been trained for tool-use using the dr-agent-lib framework. As such, running it out of the box with HuggingFace or vLLM will not work well!

See our github for more details on installation and how to run our model. Or check out our demo!

Evaluation Results

We provide evaluation instructions in our github.

Benchmark	SQAv2	HealthBench	ResearchQA	DeepResearch Bench	SimpleQA	2Wiki	WebWalker	Average
Qwen3-8B (naive rag)	40.4	16.5	56.1	33.3	52.6	18.9	8.8	32.4
Qwen3-8B (our search pipeline)	57.2	5.9	46.3	18.2	70.5	44.0	27.9	38.6
DR-Tulu-SFT-8B (this model)	72.3	38.1	68.5	39.0	75.5	66.5	31.9	56.0
DR-Tulu-8B	86.7	43.7	71.1	41.8	80.1	68.0	39.1	61.5

For more baselines, explanations of this table, and analysis of results, check out the Dr Tulu paper!

Intended uses & limitations

This model is licensed under Apache 2.0. It is intended for research and educational use in accordance with Ai2's Responsible Use Guidelines.

Training

The following hyperparameters were used during training:

learning_rate: 4e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 16
total_train_batch_size: 128
total_eval_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 5

For futher details, check out the Dr Tulu paper.

Citation

@article{shao2025dr,
  title={DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research},
  author={Shao, Rulin and Asai, Akari and Shen, Shannon Zejiang and Ivison, Hamish and Kishore, Varsha and Zhuo, Jingming and Zhao, Xinran and Park, Molly and Finlayson, Samuel G and Sontag, David and others},
  journal={arXiv preprint arXiv:2511.19399},
  year={2025}
}

rl-research/DR-Tulu-SFT-8B

DR Tulu SFT 8B

Inference and Usage

Evaluation Results

Intended uses & limitations

Training

Links

Citation

rl-research/
DR-Tulu-SFT-8B