---

base_model: huihui-ai/Huihui-Qwen3.8-27B-abliterated

library_name: gguf

pipeline_tag: image-text-to-text

license: other

language:

• en
• fr

tags:

• unsloth
• lora
• qlora
• cyber
• image-text-to-text
• vision
• video
• gguf
• llama.cpp
• ollama

model-index:

• name: Huihui-Qwen3.8-27B-abliterated-cyber — GGUF

results:

- task:

type: text-generation

name: Causal language modeling

dataset:

name: cyber SFT holdout

type: dataset_cyber.jsonl

metrics:

- type: loss

value: 0.727838

name: eval_loss

---

Huihui-Qwen3.8-27B-abliterated-cyber — GGUF

Instruction-tuned cybersecurity assistant (offensive, defensive, GRC, architecture, SOC/DFIR, RSSI).

| | |

|---|---|

| Base model | huihui-ai/Huihui-Qwen3.8-27B-abliterated |

| Domain | cyber |

| Method | LoRA / QLoRA (Unsloth) · rank 32 · α 64 |

| Quantization at train | bf16 LoRA |

| Context | 8192 tokens |

| Dataset | dataset_cyber.jsonl · train 57718 / eval 584 |

| GPU | NVIDIA RTX PRO 6000 Blackwell Server Edition (95.0 GiB) |

| Wall time | 12.8 min |

| Modalities kept | vision, video |

This checkpoint continues a strong general model and specialises it on a curated SFT corpus of cybersecurity procedures: pentest / red team, SOC and DFIR, cloud and identity, GRC (ISO, NIST, NIS2, DORA), and RSSI / project-management questions. Answers are meant to be concrete (controls, detections, hardening), not generic essays.

What changed vs the reference

Reference = the published base checkpoint huihui-ai/Huihui-Qwen3.8-27B-abliterated, plus the first in-run loss (LoRA ≈ 0 at step 0).

| Metric | Reference (base / first log) | This fine-tune | Δ |

|---|---:|---:|---:|

| Train loss (first → last logged) | 2.7575 | 0.0223 | -99.2% |

| Train loss (best) | — | 0.6722 | — |

| Eval loss (holdout, first → last) | 0.9142 | 0.7278 | -20.4% |

The first logged train loss is the closest in-run proxy for the base model (LoRA starts near zero). Option F, when executed, adds an independent holdout comparison against the frozen merged base.

Training data

• File: dataset_cyber.jsonl
• Path used at train time: /content/drive/MyDrive/finetuning/dataset_cyber.jsonl
• Split: 0.01 holdout, seed 42
• Format: chat-templated SFT (messages / instruction+output / ### Instruction + ### Response)

Training procedure

| Hyperparameter | Value |

|---|---|

| Epochs | 1 |

| Learning rate | 0.0002 |

| Warmup ratio | 0.05 |

| Device batch | 8 |

| Grad accum | 2 |

| Effective batch | 16 |

| Optim | adamw_8bit |

| Packing | True |

| LoRA targets | ['q_proj', 'k_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj', 'down_proj'] |

Intended use

Authorized defensive work, tabletop exercises, control design, detection engineering, audit readiness, and explaining attack techniques without weaponized payloads.

Out of scope: Do not use it to attack systems you do not own, to generate exploit payloads, or as a substitute for a licensed auditor or incident commander.

Multimodal

Kept towers: vision, video. Vision/audio layers were frozen during text SFT (vision=False, audio=False). Load the merged Transformers folder (or GGUF + mmproj) to keep image / video / audio.

How to use

See RUN.txt and Modelfile in this repo. Typical llama.cpp call:

llama-mtmd-cli -m Huihui-Qwen3.8-27B-abliterated-Q4_K_M.gguf --mmproj mmproj-Huihui-Qwen3.8-27B-abliterated-F32.gguf

Limitations

• Domain shift: quality drops outside the SFT topics.
• Eval above is holdout loss (and optional targeted checks). It is not a public leaderboard.
• The base model license and acceptable-use policy still apply.

License

other — inherit and respect the license of huihui-ai/Huihui-Qwen3.8-27B-abliterated.