A fine-tuned version of Qwen2.5-1.5B-Instruct specialized for biomedical tool-calling. Given a biomedical question and a set of available tool schemas, the model selects the correct tool and generates properly formatted function-call arguments.

Model Details

Property	Value
Base model	Qwen2.5-1.5B-Instruct
Fine-tuning method	QLoRA (4-bit) via Unsloth
Trainable parameters	18,464,768 (1.18% of total)
Training dataset	gxx27/BioTool (5,632 samples)
Epochs	3
Final training loss	0.20

Coverage

The model is trained on tool-calling patterns across 127 tools spanning three biomedical API families:

NCBI E-utilities (esearch, efetch, elink, BLAST, and related endpoints)
UniProt REST (protein, proteome, and taxonomy lookups)
Ensembl REST (coordinate mapping, variant effect prediction, comparative genomics)

This covers genomics, proteomics, and comparative biology tool-use. It does not cover clinical-facing APIs such as ICD-10 lookup, drug databases, or clinical trial registries.

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Rumiii/Qwen-BioTool-1.5B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

tools = [
    {
        "type": "function",
        "function": {
            "name": "esearch",
            "description": "Search an NCBI Entrez database and return UIDs matching a text query.",
            "parameters": {
                "type": "object",
                "properties": {
                    "db": {"type": "string", "description": "Entrez database name"},
                    "term": {"type": "string", "description": "Search query"},
                },
                "required": ["db", "term"],
            },
        },
    }
]

messages = [{"role": "user", "content": "Search PubMed for articles on BRCA1 mutations."}]

inputs = tokenizer.apply_chat_template(
    messages,
    tools=tools,
    add_generation_prompt=True,
    tokenize=True,
    return_tensors="pt",
    return_dict=True,
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

Known Limitations

The arguments field in generated tool calls is a JSON-encoded string, not a nested JSON object. Downstream code should apply an additional json.loads() to it before use.
Trained exclusively on tool-calling examples; conversational ability is inherited from the base model rather than reinforced during fine-tuning.
Tool coverage is limited to NCBI, UniProt, and Ensembl. Queries requiring other biomedical or clinical APIs are outside its trained scope.
Not intended for clinical decision-making or diagnostic use.

Training Data

BioTool: a biomedical function-calling dataset of 7,040 human-verified query-to-API-call pairs across NCBI, UniProt, and Ensembl.

bibtex

@misc{gao2026biotoolcomprehensivetoolcallingdataset,
      title={BioTool: A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language Models},
      author={Xin Gao and Ruiyi Zhang and Meixi Du and Peijia Qin and Pengtao Xie},
      year={2026},
      eprint={2605.05758},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2605.05758},
}

This model was trained with Unsloth and Hugging Face's TRL library.

License

Apache 2.0