Downloads · 30 days
443
18% of all-time downloads
mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7-MLX-VLM-4bit
Qwen3.8-27B-Uncensored-bughunter-v7-MLX-VLM-4bit is a image-text-to-text model from mmp2055. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
MLX 4-bit build optimized for Apple Silicon, converted from mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7 (GGUF version).
Downloads · 30 days
443
18% of all-time downloads
All-time downloads
2.4K
Public
Parameters
27.4B
16.1 GB on disk
Likes
2
Public
Click a slice to open those files.
.safetensors16.1 GB · 100%
How the weights are stored.
U3226.9B · 98%
From the Hugging Face model README
MLX 4-bit build optimized for Apple Silicon, converted from mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7 (GGUF version).
Same fine-tune, same dataset, same LoRA — repackaged as MLX safetensors 4-bit for native Metal acceleration on M-series Macs. Ships with the multimodal vision tower + projector intact, so image understanding works out of the box via mlx-vlm.
For non-Apple platforms, use the GGUF version.
mlx-vlm runtime detects text tower + vision tower + projector automatically from config.json.Runs on:
mlx-vlm CLI and Python API — the recommended runtime for VLM on Apple Silicon.mlx-vlm (custom servers, chat wrappers, etc.).Requires: macOS on Apple Silicon (M1/M2/M3/M4/M5). Not compatible with Linux, Windows or Intel Macs — for those, use the GGUF build.
Measured on Apple M5 with 32 GB unified memory:
| Config | Peak memory | Text tok/s | With image |
|---|---|---|---|
| Q4 (this build), 32k context | ~16 GB | 6-9 | +2-5 s per image (first-pass encoding) |
For a full comparison and honest trade-offs, see the notes below. On unified memory of 48 GB or more, this model runs at longer contexts (64k+) with the same speed profile.
Covers the full lifecycle of a bug bounty engagement:
{{7*7}} on Jinja/Twig/Freemarker/Velocity), RCE via deserialization, XXE in SAML/SOAP/SVG/OOXML.The MLX-VLM build preserves the multimodal capability of the base model. Practical uses in bug bounty context:
Note: the vision tower comes from the base model and was not fine-tuned — the added value is on the text side. Image understanding is general-purpose (no security-specific visual training).
When a system prompt is provided, the model incorporates ethical research rules:
whoami/id for RCE; SELECT version() for SQLi; two owner accounts for IDOR.This MLX build is a quantization + repackaging step on top of the base fine-tune. The full training details (dataset, hyperparameters, hygiene) are documented in the GGUF v7 model card.
AutoModelForImageTextToText.from_pretrained() (multimodal class, preserves vision tower + projector) in bfloat16. LoRA adapter applied via PeftModel.from_pretrained() and merged with model.merge_and_unload(). Full merged model saved (~55 GB with vision).mlx_vlm.convert -q --hf-path <merged> --mlx-path <mlx> performs the MLX quantization to 4-bit. Result: ~15 GB across 3 safetensors shards, 4.695 bits per weight including the vision tower.system messages (compatibility with IDE agents that inject their own system prompt on top of one the user set).Full pipeline (Colab notebook + Mac conversion script) is documented at the source. To reproduce this exact MLX build:
mlx_vlm.convert -q steps above.mlx-vlm (recommended)Install once:
pipx install mlx-vlm
Text-only:
mlx_vlm.generate \
--model mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7-MLX-VLM-4bit \
--prompt "¿qué es un IDOR?" \
--max-tokens 300
With an image (bug bounty screenshot use case):
mlx_vlm.generate \
--model mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7-MLX-VLM-4bit \
--image screenshot-waf-block.png \
--prompt "¿qué WAF es y qué payloads probarías para bypass?" \
--max-tokens 500
As an OpenAI-compatible server:
mlx_vlm.server --model mmp2055/Qwen3.8-27B-Uncensored-bughunter-v7-MLX-VLM-4bit --port 8080
Endpoint at http://127.0.0.1:8080/v1/chat/completions with the OpenAI vision message format (type: image_url blocks).
~/.lmstudio/models/<publisher>/<repo>/.Runtime Extensions → MLX).qwen35 does not tolerate KV quantization well).Text inference works natively. Vision inference in the LM Studio GUI depends on the version supporting MLX-VLM sidecars — check your build. If GUI vision isn't available, use mlx_vlm.generate from the CLI.
You can use the model without a system prompt (it behaves correctly by default). For bug bounty tasks specifically, the recommended prompt is:
You are a professional bug bounty researcher. You work on HackerOne, Bugcrowd, Intigriti, YesWeHack, Immunefi and independent programs.
Inviolable rules:
- In-scope assets only, verified against the program scope before every request.
- Sustained throughput ≤1-2 req/s.
- Minimal non-destructive PoC.
- OOB callbacks always to owned infrastructure.
- Never use credentials or tokens found in a finding to validate access.
- Minimum reportable severity: High or Critical.
Answer in Spanish. Payloads, commands and report templates in English.
This model is distributed for:
It is not distributed for:
The user assumes all legal responsibility for the use of the model.
qwen35, hybrid attention with Gated DeltaNet + full attention, extended context up to 262,144 tokens, Multi-Token Prediction, native vision tower).qwen35.mlx-vlm — the Apple Silicon-native VLM runtime.@misc{qwen38_bughunter_v7_mlx,
title = {Qwen3.8-27B-Uncensored-bughunter-v7 (MLX 4-bit with vision): Apple Silicon build of the bug bounty fine-tune},
year = {2026},
base = {orcarouter/Qwen3.8-27B-Uncensored},
method = {LoRA fine-tune (rank=8, lr=1e-5, 1 epoch, QLoRA 4-bit, Unsloth) + mlx_vlm.convert -q},
data = {14914 ShareGPT examples distilled from public bug bounty corpus}
}