HuntressQwen3 is an uncensored, agentic fine-tune of Qwen3-14B designed specifically for penetration testing, bug bounty hunting, and autonomous terminal agent workflows (such as OpenCode, VSCode Cline, and Roo-Code).
When asked who it is, it asserts:
"I am HuntressQwen3, an uncensored model built for hacking, penetration testing and bug bounty."
<think>)Unlike standard LLMs that either hallucinate immediately or output boilerplate warnings, HuntressQwen3 uses distilled 100% English internal reasoning traces:
<think>.HuntressQwen3 is fine-tuned on multi-turn ReAct (Reason → Action → Observation → Pivot) cycles:
curl, ffuf, subfinder, nmap) rather than generic conceptual advice.dbms_pipe.receive_message(('a'), 5) and conditional FROM dual). Eliminates the common LLM hallucination of using MySQL SLEEP() on Oracle.pg_sleep), MSSQL (WAITFOR DELAY), and MySQL (SLEEP).<script> and onerror= are blocked, it strictly adheres to the rule and outputs alternative event handlers (<svg onload=...>, <details ontoggle=...>, autofocus attributes) with 0 forbidden tokens.| Test Vector | Base Qwen3-14B | Huntress v3 | Huntress v4 | HuntressQwen3 v5 (Final) |
|---|---|---|---|---|
| Thinking Language | Mixed CJK | Contaminated | CJK (起来...) |
🟢 100% Pure English (0 CJK tokens) |
| Oracle Blind SQLi | MySQL SLEEP() | Length only | MySQL SLEEP() | 🟢 Accurate dbms_pipe + dual |
| Constraint Following (WAF) | ❌ Violated | ❌ Violated | 🟡 Partial | 🟢 100% Strict Pass |
| Real Target Pentest | Fabricated | Repetition loop | Hallucinated code | 🟢 Actionable CLI Recon (ffuf/curl) |
| Agentic Loop (OpenCode) | Static text | Static text | Static text | 🟢 Autonomous ReAct Decisions |
Repetition Loops (maxdup) |
High | Infinite loops | 1 | 🟢 1 (Completely Eliminated) |
| Training Loss | N/A | 1.140 | 0.938 | 🟢 0.860 (Best Fit) |
Download the model weights from Hugging Face:
ollama run hf.co/ArunShekhar/HuntressQwen3-14B-UNCENSORED-agentic
Or build locally with the included Modelfile:
ollama create HuntressQwen3 -f Modelfile
ollama run HuntressQwen3
To toggle reasoning via Ollama API:
curl http://127.0.0.1:11434/api/chat -d '{
"model": "HuntressQwen3",
"think": true,
"messages": [{"role": "user", "content": "Reflected XSS in search parameter, WAF blocks <script> and onerror=. Give payloads."}]
}'
(Set "think": false for fast direct mode).
In your agent client (~/.local/state/opencode/model.json or Cline settings):
OllamaHuntressQwen3:latest16,384The model will autonomously plan recon steps, emit terminal execution commands, parse returned tool responses, and escalate attacks step-by-step.
llama-cli -m HuntressQwen3-Q4_K_M.gguf -c 16384 -cnv --jinja
| File | Size | Description |
|---|---|---|
HuntressQwen3-Q4_K_M.gguf |
8.4 GB | Standalone merged quantized model (Download on Hugging Face) |
HuntressQwen3-lora-f16.gguf |
246 MB | Trained LoRA adapter in f16 GGUF format |
Modelfile |
3.8 KB | Ollama configuration with agentic rules and tuned sampling parameters |
run_huntress.py |
1.6 KB | Python interactive terminal client |
NOTICE & LICENSE |
Text | Apache-2.0 license and upstream attribution |
0.40.85201.11.1516,384HuntressQwen3 is an uncensored model developed strictly for authorized security assessments, penetration testing under contractual agreements, bug bounty hunting following program guidelines, and academic security research.
The author and contributors assume no liability for unauthorized or malicious use. Always ensure you have explicit written authorization before conducting security testing against any application, server, or network infrastructure.
Qwen/Qwen2.5-14B