LiquidAI/LFM2-2.6B-Longevity

Docs • Discord

Longevity-LLM - LFM2-2.6B

Longevity-LLM (L-LLM) is a family of compact, domain-adapted language models for interpreting heterogeneous aging biology data. Longevity LFMs are available in two sizes:

• LFM2-1.2B-Longevity
• LFM2-2.6B-Longevity (this model)

This checkpoint, L-LFM2-2.6B, was produced by full-parameter supervised fine-tuning of LiquidAI/LFM2-2.6B on aging-related multi-omics and clinical data.

The family was developed jointly by Insilico Medicine and Liquid AI and accompanies the study "An Open Benchmark and Language Models for AI in Aging Biology" (Zhavoronkov et al., 2026).

Model details

• Base model: LiquidAI/LFM2-2.6B
• Architecture: Hybrid Liquid model with multiplicative gates and short convolutions.
• Context length: 32,768 tokens
• Language: English

Training data. The model was trained on the shared L-LLM corpus spanning aging biology. See LongevityBench for more details.

Training procedure. L-LFM2-2.6B was trained with full-parameter supervised fine-tuning. Prompts were formatted in ChatML with a dynamic-thinking template (user turns suffixed with /think or /no_think to select response mode at inference).

Chat Template

LFM2 uses a ChatML-like format. See the Chat Template documentation for details. Example:

<|startoftext|><|im_start|>system
You are a biomedical AI specialized in aging biology, trained on genomic, proteomic, and clinical data.<|im_end|>
<|im_start|>user
What is C. elegans?<|im_end|>
<|im_start|>assistant

You can use tokenizer.apply_chat_template() to format your messages automatically.

Inference

LFM2 is supported by many inference frameworks. See the Inference documentation for the full list.

Name	Description	Docs	Notebook
Transformers	Simple inference with direct access to model internals.	Link
vLLM	High-throughput production deployments with GPU.	Link
SGLang	High-throughput production deployments with GPU.	Link	—
llama.cpp	Cross-platform inference with CPU offloading.	Link
MLX	Apple's machine learning framework optimized for Apple Silicon.	Link	—
LM Studio	Desktop application for running LLMs locally.	Link	—

Quick start with Transformers (compatible with transformers>=5.1.0)

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model_id = "LiquidAI/LFM2-2.6B-Longevity" 
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    dtype="bfloat16",
)
tokenizer = AutoTokenizer.from_pretrained(model_id)

# Generate answer
messages = [
    {"role": "system", "content": "You are a biomedical AI specialized in aging biology, trained on genomic, proteomic, and clinical data."},
    {"role": "user", "content": "What are the hallmarks of aging?"},
]

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

output = model.generate(
    input_ids,
    do_sample=True,
    temperature=0.3,
    min_p=0.15,
    repetition_penalty=1.05,
    max_new_tokens=1500
)

print(tokenizer.decode(output[0], skip_special_tokens=False))

Intended use and limitations

Intended for research on aging biology and omics interpretation. Outputs are model predictions, not clinical advice, and should be validated experimentally. Performance is strongest on the modalities represented in the training corpus.

Contact

• Got questions or want to connect? Join our Discord community
• If you are interested in custom solutions with edge deployment, please contact our sales team.

Citation

@misc{liquid_ai_2026,
    author       = { Liquid AI },
    title        = { LFM2-2.6B-Longevity (Revision 601e34d) },
    year         = 2026,
    url          = { https://huggingface.co/LiquidAI/LFM2-2.6B-Longevity },
    doi          = { 10.57967/hf/9888 },
    publisher    = { Hugging Face }
}

@article{liquidai2025lfm2,
 title={LFM2 Technical Report},
 author={Liquid AI},
 journal={arXiv preprint arXiv:2511.23404},
 year={2025}
}