Instructions to use Volpidash/deepseek-r1-14b-security with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Volpidash/deepseek-r1-14b-security with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-R1-Distill-Qwen-14B") model = PeftModel.from_pretrained(base_model, "Volpidash/deepseek-r1-14b-security") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Volpidash/deepseek-r1-14b-security with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf Volpidash/deepseek-r1-14b-security:Q4_K_M # Run inference directly in the terminal: llama cli -hf Volpidash/deepseek-r1-14b-security:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Volpidash/deepseek-r1-14b-security:Q4_K_M # Run inference directly in the terminal: llama cli -hf Volpidash/deepseek-r1-14b-security:Q4_K_M
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf Volpidash/deepseek-r1-14b-security:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf Volpidash/deepseek-r1-14b-security:Q4_K_M
Use Docker
docker model run hf.co/Volpidash/deepseek-r1-14b-security:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use Volpidash/deepseek-r1-14b-security with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Volpidash/deepseek-r1-14b-security" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Volpidash/deepseek-r1-14b-security", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Volpidash/deepseek-r1-14b-security:Q4_K_M
- Ollama
How to use Volpidash/deepseek-r1-14b-security with Ollama:
ollama run hf.co/Volpidash/deepseek-r1-14b-security:Q4_K_M
- Unsloth Desktop
- Docker Model Runner
How to use Volpidash/deepseek-r1-14b-security with Docker Model Runner:
docker model run hf.co/Volpidash/deepseek-r1-14b-security:Q4_K_M
- Lemonade
How to use Volpidash/deepseek-r1-14b-security with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Volpidash/deepseek-r1-14b-security:Q4_K_M
Run and chat with the model
lemonade run user.deepseek-r1-14b-security-Q4_K_M
List all available models
lemonade list
- Atomic Chat
Download benchmark-protocol.md from Volpidash/deepseek-r1-14b-security: direct link, hf CLI and curl.
- Browser
- Download file 4.37 kB
-
https://huggingface.co/Volpidash/deepseek-r1-14b-security/resolve/main/benchmark-protocol.md
- Command line
-
hf download hf://Volpidash/deepseek-r1-14b-security/benchmark-protocol.md
-
curl -L -o benchmark-protocol.md https://huggingface.co/Volpidash/deepseek-r1-14b-security/resolve/main/benchmark-protocol.md
Security benchmark protocol
Protocol: balanced-letter-ranking-v1; frozen selection seed: 20261109.
| Suite | Questions | Correct answer positions |
|---|---|---|
| Held-out Pentesting Explanations | 176 | A: 44, B: 44, C: 44, D: 44 |
| SecEval adapted subset | 200 | A: 50, B: 50, C: 50, D: 50 |
Scoring
Each question is shown with four choices using the base model's chat template. The reasoning prefix is closed before scoring. The evaluator compares the next-token logits for A, B, C, and D, and selects the highest. Each label is verified to be a single token in its actual prompt context. This measures forced direct-answer ranking, with no generated reasoning.
The base is evaluated by disabling the trained LoRA adapter; the fine-tune enables that same adapter. Both use the same NF4 double-quantized weights, FP16 compute, prompt, GPU, and software. Dropout is disabled. Maximum prompt length is 768 tokens. Base/adapter evaluation order alternates. Conditional A–D probabilities and the probability mass of those four tokens within the full vocabulary are recorded.
Each suite has a 25% uniform-guessing expectation and 25% majority-label baseline. Reported accuracy includes descriptive Wilson 95% intervals. Paired corrected/regressed counts and topic breakdowns are provided. Topics overlap, and a small or dependent subset does not support broad population claims.
Selection and overlap checks
Pentesting questions come from the held-out split. SecEval selection excludes invalid or multiple-answer cases and uses a seeded round-robin sample across primary topics. Questions with detected position-dependent options, matching training question text/CVE IDs, a training-stem word-4-gram Jaccard similarity of at least 0.8, or overlength prompts are excluded. These lexical checks do not detect every semantic duplicate or base-model pretraining contamination.
Choices are reordered to balance labels. The frozen manifest records source IDs, zero-based original choice indices in their new order, answer labels, topics, prompt lengths, and hashes. It does not redistribute question text. Full question text remains local. Selection counts and exclusions are in the manifest.
Sources
- SecEval by XuanwuAI: evaluation-only dataset, CC BY-NC-SA. See the authors' project. The adapted single-answer subset and this scoring method are not official SecEval leaderboard results. The dataset's terms are separate from the model's MIT license.
- Pentesting Explanations by Christopher Dickinson / theelderemo: Apache-2.0. Synthetic answers have not been independently fact-checked.
Execution and reproducibility
The evaluator runs automatically after the training process exits, before CPU merge/export, so it does not compete with training for VRAM. Per-question paired checkpoints allow resumption only with identical model, adapter, suite, scorer, and library fingerprints. Results contain adapter SHA-256, pinned base revision, software versions, device, and per-question predictions without question text.
Published scripts are under evaluation/. To reproduce in the original workspace layout, place them in deepseek14b/ alongside model_config.json, base/, data/train.jsonl, data/eval.jsonl, and outputs/adapter/. Fetch the pinned source datasets in the manifest; place SecEval's questions.json in data/benchmark-sources/SecEval/. Run benchmark_data.py, then run_benchmarks.py --validate-only, then run_benchmarks.py after training. Verify the generated manifest hashes against this release. The corpus hashes are provided for verification; the corpus and SecEval question text are not bundled with these reports.
Scope
Scores apply to the NF4 runtime, not to the Q4_K_M GGUF. They are not generated-answer pass@1 or measurements of normal R1 reasoning, tool calling, exploit discovery, or current CVE completeness. No held-out iOS 27 advisory score is available. The additional 10 generated probes are diagnostics, not a validated factual-accuracy benchmark.