Fastino Nemotron 3.5 Lightning Finance

Fastino-Nemotron-3.5-Lightning-Finance is a 30B-parameter, 3B-active mixture-of-experts model specialized for financial reasoning, extraction, and research fine-tuned on LoRA with the Fastino Fine-Tuning Agent.

• Developed by: Fastino Labs, in collaboration with NVIDIA
• Base checkpoint: NVIDIA Nemotron 3.5 Lightning, July 29, 2026 release
• Post-training checkpoint: Fastino-Finance
• License: Apache 2.0
• Language: English
• Modalities: Text
• Model Release Blog: Release Blog
• Fine-Tuning Agent: Private Preview
• Research: Fastino Fine-Tuning Agent

What it is designed for

The model targets financial document reasoning, numerical question answering over filings and tables, numeric span extraction, financial entity recognition, conversational analysis, and source-grounded financial research. The evaluation suite includes FinQA, TAT-QA, SEC-Num, FinEntity, BizFinBench, BigFinanceBench, ConvFinQA, and FiQA.

Quickstart

The published weights are BF16 and require about 66 GB before runtime overhead. An 80 GB or larger GPU, or tensor parallelism across multiple GPUs, is recommended.

pip install "vllm==0.23.0"

from vllm import LLM, SamplingParams

model_id = "fastino/Fastino-Nemotron-3.5-Lightning-Finance"

llm = LLM(
    model=model_id,
    trust_remote_code=True,
    dtype="bfloat16",
    max_model_len=4096,
)

outputs = llm.generate(
    ["Check this financial calculation and explain the result: ..."],
    SamplingParams(temperature=0.0, max_tokens=512),
)
print(outputs[0].outputs[0].text)

Post-training recipe

The Fastino Fine-Tuning Agent autonomously built evaluation sets, curated data, explored training mixtures and hyperparameters, recovered failed experiments, evaluated transfer, and selected the final checkpoint.

The winning adapter was trained on 13,698 de-duplicated examples covering:

• financial document calculation and executable reasoning;
• hybrid text-and-table question answering;
• business-finance reasoning across calculation, extraction, temporal reasoning, prediction, and knowledge tasks; and
• numerical span extraction from SEC disclosures;
• financial entity extraction;
• source-grounded financial research trajectories.

The final mix deliberately allocated substantial coverage to SEC-Num and BizFinBench while retaining FinQA, TAT-QA, FinEntity, and BigFinanceBench examples. It was trained from the base checkpoint for two epochs with LoRA rank 32, learning rate 1e-4, and sequence packing disabled. Exact duplicates were removed within each source before mixture construction. Evaluation examples and labels were excluded from training.

Benchmark evaluation

Base and fine-tuned scores below use the Nemotron-3.5-Lightning July 29 checkpoint, inputs, prompts, decoding settings, inference route, and evaluator for each row.

In-domain benchmarks

Transfer to unseen benchmarks

Performance on related tasks the model was not explicitly trained for.

Evaluation protocol

The agent used development evaluations for experiment selection and a separately frozen held-out lane for final characterization. Score-bearing comparisons were accepted only when base and candidate shared the same examples, prompt construction, decoding parameters, serving route, evaluator identity, and aggregation. Training mixtures were hashed and de-duplicated, and benchmark evaluation rows were excluded from training.

Limitations

This is a specialized model whose capabilities are best characterized by the tasks above. Performance outside these tasks has not been comprehensively evaluated. Outputs are not a substitute for professional financial advice; high-stakes use requires independent validation and qualified human oversight.

Fine-tuning

To fine-tune or serve models like this one without managing GPUs, see the Fastino API, which offers hosted training and inference.

Citation

@misc{atreja2026pioneeragentcontinualimprovement,
      title={Pioneer Agent: Continual Improvement of Small Language Models in Production},
      author={Dhruv Atreja and Julia White and Nikhil Nayak and Kelton Zhang and Henrijs Princis and George Hurn-Maloney and Ash Lewis and Urchade Zaratiana},
      year={2026},
      eprint={2604.09791},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2604.09791},
}

License

This model is licensed under the Apache License 2.0.

Links

• Fastino Labs
• Fine-Tuning Agent private preview
• Supporting research
• NVIDIA Nemotron 3.5 Lightning base model