---
license: apache-2.0
tags:
  - cybersecurity
  - document-classification
  - gguf
  - ollama
  - qwen
  - lora
base_model: Qwen/Qwen3.5-27B
---

# TorchSight Beam f16

Cybersecurity document classifier. LoRA fine-tune of **Qwen 3.5 27B**, full
half-precision (no quantization). Approximately 53 GB GGUF.

Recommended hardware: 96 GB+ GPU. Use this variant for research / reference;
for production deployment, prefer q4_K_M (default) or q8_0.

Released alongside:

> Dobrovolskyi, I. *Security Document Classification with a Fine-Tuned Local
> Large Language Model: Benchmark Data and an Open-Source System.* Journal of
> Information Security and Applications, 2026.

## Benchmark results

Evaluated under identical methodology (alpaca prompt, Ollama `/api/generate`,
temperature = 0, `num_predict = 2048`) on the companion dataset
[`torchsight/cybersecurity-classification-benchmark`](https://huggingface.co/datasets/torchsight/cybersecurity-classification-benchmark).
Canonical numbers live in that repo's `BENCHMARK_NUMBERS.md`.

### Primary — eval-1000-synthetic (n = 1,000)

| Model              | Type             | Cat. acc [95% CI]      | Subcat. acc |
|---|---|---:|---:|
| Beam q4_K_M        | Local (LoRA)     | 95.0% [93.5, 96.2]     | 48.2% |
| **Beam f16**       | Local (LoRA)     | **93.2%** [91.5, 94.6] | **51.1%** |
| Beam q8_0          | Local (LoRA)     | 93.0% [91.2, 94.4]     | 51.4% |
| Claude Sonnet 4    | Commercial API   | 79.9% [77.3, 82.3]     | 23.0% |
| Claude Opus 4      | Commercial API   | 79.9% [77.3, 82.3]     | 22.5% |
| GPT-5              | Commercial API   | 76.9% [74.2, 79.4]     | 11.6% |
| Gemini 2.5 Pro     | Commercial API   | 75.4% [72.6, 78.0]     | 21.0% |
| Qwen 3.5 27B base  | Local (no LoRA)  | 86.3% [84.0, 88.3]     | 19.0% |
| Regex (48 patterns)| Rule-based       | 52.7% [49.6, 55.8]     | —     |

### External — eval-500-external (n = 500)

| Model              | Cat. acc [95% CI]      | Δ vs. primary |
|---|---:|---:|
| Beam q4_K_M        | 93.8% [91.3, 95.6]     | −1.2 pp |
| **Beam f16**       | **91.2%** [88.4, 93.4] | **−2.0 pp** |
| Beam q8_0          | 91.2% [88.4, 93.4]     | −1.8 pp |
| Claude Sonnet 4    | 86.4% [83.1, 89.1]     | +6.5 pp |
| Gemini 2.5 Pro     | 82.0% [78.4, 85.1]     | +6.6 pp |
| Qwen 3.5 27B base  | 86.6% [83.3, 89.3]     | +0.3 pp |
| GPT-5              | 65.8% [61.5, 69.8]     | −11.1 pp |
| Regex baseline     | 29.6% [25.8, 33.7]     | −23.1 pp |

## Usage with Ollama

```bash
ollama pull torchsight/beam-f16
ollama run torchsight/beam-f16
```

Or via the [TorchSight CLI](https://github.com/IvanDobrovolsky/torchsight).

## Training

- Base: Qwen 3.5 27B (dense)
- Method: LoRA (r = 128, α = 256), bf16, 5 epochs
- Dataset: 78,358 balanced samples — see [`torchsight/beam-training-data`](https://huggingface.co/datasets/torchsight/beam-training-data)
- Hardware: 8× NVIDIA A100 80GB SXM4, 10.5 hours

## License

Apache 2.0. The base model (Qwen 3.5 27B) carries its own license; consult
upstream terms for use.