# secgpt

SecGPT 是由 云起无垠 于 2023 年正式推出的开源大模型，专为网络安全场景打造，旨在以人工智能技术全面提升安全防护效率与效果。

- **Kind:** model (Hugging Face)
- **Hugging Face:** https://huggingface.co/clouditera/secgpt
- **Owner:** clouditera (https://forefy.com/u/hf/clouditera)
- **Parameters:** 14.8B
- **Context:** 32768 tokens
- **License:** apache-2.0
- **Focus:** defensive
- **Architecture:** Qwen2ForCausalLM
- **Pinned revision:** `b742806ac09f57d9b94893d2d381680fc052d737`
- **Page:** https://forefy.com/models/f44167a6-0b0d-55bf-b8df-cc6b3d9262eb
- **API (JSON):** https://forefy.com/api/asr/models/f44167a6-0b0d-55bf-b8df-cc6b3d9262eb

## What was done to it

- finetune (declared evidence, from Qwen/Qwen2.5-14B-Instruct)

## Lineage

Qwen/Qwen2.5-14B → Qwen/Qwen2.5-14B-Instruct → clouditera/secgpt

## Checks (repo at the pinned revision; not a behavior test)

- **Identity** [pass]: No lookalike name or conflicting provenance
- **Weights format** [pass]: safetensors only, nothing runs on load
- **Pickle files** [pass]: None.
- **Remote code** [info]: 1 Python file(s), not loaded automatically
- **Chat template** [pass]: Same as Qwen/Qwen2.5-14B-Instruct's
- **Refusals modified** [pass]: No
- **License** [pass]: apache-2.0
- **Lineage** [pass]: Derives from Qwen/Qwen2.5-14B-Instruct
- **Model card** [warn]: 3 of 6, missing training data, evaluation results, intended use and limitations
- **Safetensors headers** [pass]: Every byte is indexed, no hidden data
- **Tensor layout** [pass]: Same tensors and shapes as Qwen/Qwen2.5-14B-Instruct

## Variants

- F16 safetensors (safetensors, 29.5 GB)

## Run it

- transformers: `AutoModelForCausalLM.from_pretrained("clouditera/secgpt", revision="b742806ac09f57d9b94893d2d381680fc052d737")`