What license applies to SC117/QwenPaw-Flash-9B-heretic-GGUF?

License: apache-2.0. Verify terms on Hugging Face before commercial use.

How do I run SC117/QwenPaw-Flash-9B-heretic-GGUF locally?

Download a GGUF file from this page and load it in guIDE or llama.cpp. Pipeline task: text-generation.

Model Intelligence Sheet

SC117/QwenPaw-Flash-9B-heretic-GGUF overview

language: en zh base model: Qwen/Qwen3.5 9B agentscope ai/QwenPaw Flash 9B tags: heretic abliteration uncensored qwen3.5 gguf benchlocal benchmark agent tool c…

ggufhereticabliterationuncensoredqwen3.5benchlocalbenchmarkagenttool-callenzhbase_model:Qwen/Qwen3.5-9Bbase_model:quantized:Qwen/Qwen3.5-9Blicense:apache-2.0endpoints_compatibleregion:usconversational

Runs locally from ~879.0 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads

6,834

Likes

Pipeline

—

Author

SC117

Repository Files & Downloads

5 GGUF files detected

Direct downloads for local inference

File	Type	Quantization	Size	Link
QwenPaw-Flash-9B-heretic-F16.gguf	GGUF	F16	16.69 GB	Download
QwenPaw-Flash-9B-heretic-Q4_K_M.gguf	GGUF	Q4_K_M	5.24 GB	Download
QwenPaw-Flash-9B-heretic-Q6_K.gguf	GGUF	Q6_K	6.85 GB	Download
QwenPaw-Flash-9B-heretic-Q8_0.gguf	GGUF	Q8_0	8.87 GB	Download
mmproj-QwenPaw-Flash-9B-heretic-BF16.gguf	GGUF	BF16	879.0 MB	Download

Model Details

Model ID	SC117/QwenPaw-Flash-9B-heretic-GGUF
Author	SC117
Pipeline	—
License	apache-2.0
Base model	Qwen/Qwen3.5-9B,agentscope-ai/QwenPaw-Flash-9B
Last modified	2026-10-03T10:21:42.000Z

Model README

---

language:

base_model:

Qwen/Qwen3.5-9B
agentscope-ai/QwenPaw-Flash-9B

tags:

heretic
abliteration
uncensored
qwen3.5
gguf
benchlocal
benchmark
agent
tool-call

license: apache-2.0

---

</div>

<h1 style="margin: 0 0 8px 0; font-size: 32px; font-weight: 700; color: #1d1d1f; letter-spacing: -0.5px; border: none; position: relative; z-index: 1;">QwenPaw-Flash-9B-heretic</h1>

<p style="margin: 8px 0 0 0; font-size: 14px; position: relative; z-index: 1;"><span style="color: #86868b;">English</span> | <a href="https://huggingface.co/SC117/QwenPaw-Flash-9B-heretic-GGUF/blob/main/README_zh.md" style="color: #007aff; text-decoration: none;">📖 中文文档</a></p>

</div>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;">QwenPaw-Flash-9B-heretic MTP Version: <a href="https://huggingface.co/SC117/QwenPaw-Flash-9B-heretic-MTP-GGUF" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">[QwenPaw-Flash-9B-heretic-MTP-GGUF</a> ]</p>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;"><b>🏆 BenchLocal Total: 3873/5000 (77.5%) — Outperforms 35B MoE models</b><br></p>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;"><i>Uncensored · Abliterated · Agent-Optimized</i></p>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;">Uncensored version of <b>QwenPaw-Flash-9B</b>, processed with <a href="https://github.com/p-e-w/heretic" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">Heretic</a> v1.3.0 abliteration. Fine-tuned from Qwen3.5-9B for autonomous agent scenarios.</p>

</div>

<div style="background: linear-gradient(135deg, #ff6b35 0%, #f7931e 100%); padding: 12px 16px; color: white; font-weight: 700; font-size: 14px; display: flex; align-items: center; gap: 8px;"><span>📊</span> 🏆 BenchLocal Benchmarks</div>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #64748b; font-style: italic; border-left: 3px solid #ffedd5; padding-left: 12px;"><b>Test Environment</b>: NVIDIA RTX 5070 Ti (16GB) · llama.cpp (turboquant build) · Q6_K quant</p>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #64748b; font-style: italic; border-left: 3px solid #ffedd5; padding-left: 12px;"><b>Framework</b>: <a href="https://github.com/stevibe/benchlocal" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">BenchLocal</a> — local model agent evaluation suite</p>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #64748b; font-style: italic; border-left: 3px solid #ffedd5; padding-left: 12px;"><b>Methodology</b>: Each scenario run <b>once</b>, no retries, no second attempts</p>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Benchmark</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Score</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Accuracy</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Results</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;"><b>ToolCall-15</b> 🛠️</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;"><b>HermesAgent-20</b> 🤖</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;"><b>BugFind-15</b> 🐛</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;"><b>Total</b></td>

</tr>

</tbody></table>

<h3 style="margin: 16px 0 8px 0; font-size: 14px; color: #1e293b; font-weight: 700;">🛠️ ToolCall-15 — Tool Calling Stability (93.3%)</h3>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Result</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Score</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Scenario</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Simple tool call</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Multi-parameter tool</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Nested tool call</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Parameter type conversion</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Relative date/time parsing</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Optional parameter handling</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Error return value handling</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Chained calls</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Batch parameters</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">JSON parameter parsing</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Prompt boundary</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Ambiguous request rejection</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Retry mechanism</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">State persistence</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Concurrent safety</td>

</tr>

</tbody></table>

<h3 style="margin: 16px 0 8px 0; font-size: 14px; color: #1e293b; font-weight: 700;">🤖 HermesAgent-20 — Complex Agent Tasks (77.2%)</h3>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;"><b>4-Model Comparison (9B vs 35B / 26B):</b></p>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Model</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Score</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">⚠️Partial</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;"><b>🐾 QwenPaw 9B (ours)</b></td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">🧠 Qwen3.6 35B A3B Thinking ON</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">🔮 Gemma 4 26B A4B</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">⚡ Qwen3.6 35B A3B Thinking OFF</td>

</tr>

</tbody></table>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Result</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Score</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Scenario</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Replace contradictory memory</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Memory near capacity</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Reject malicious injection</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Cross-session recall</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Fix a real failing test</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Background process management</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Programmatic tool chaining</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Browser automation</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Create a skill</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Discover an existing skill</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Patch an existing skill</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Manage skill files</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Create a cron job</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Update a cron job</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Trigger a cron job</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Send a message</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Parallel delegation</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Delete a target</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Recover and retry</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Ambiguous destructive request</td>

</tr>

</tbody></table>

<h3 style="margin: 16px 0 8px 0; font-size: 14px; color: #1e293b; font-weight: 700;">🐛 BugFind-15 — Code Debugging (61.9%)</h3>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Result</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Score</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Scenario</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Python syntax error</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">JavaScript closure</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Mostly correct</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">False positive</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Null pointer exception</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Go loop variable capture</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Partial identification</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">JS Promise/Await</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Root cause missed</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">SQL injection vulnerability</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Python memory leak</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">C++ segfault</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Misleading code</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">False positive</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Silent invalid input</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct direction, incomplete</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Complex scenario</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Timeout (300s)</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Race condition</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Correct fix</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Production data missing</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Near perfect</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Concurrent data race</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Imprecise localization</td>

</tr>

</tbody></table>

</div>

<ul style="margin: 0 0 12px 0; padding-left: 20px;"><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Base model: <a href="https://huggingface.co/agentscope-ai/QwenPaw-Flash-9B" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">QwenPaw-Flash-9B</a> (Qwen3.5-9B fine-tuned for QwenPaw autonomous agent scenarios)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Tool: Heretic v1.3.0 (automatic directional ablation)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Best trial**: #194 / 230 trials</li></ul>

</div>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Metric</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Original</th>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">After Heretic</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Refusal rate (100 prompts)</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">KL divergence</td>

</tr>

</tbody></table>

</div>

<p style="margin: 0; font-family: monospace; background: #f8fafc; padding: 10px 14px; border-radius: 6px; border: 1px solid #e2e8f0; font-size: 12px; color: #1e293b; white-space: pre-wrap;">direction_index = 21.13

attn.o_proj.max_weight = 1.42

attn.o_proj.max_weight_position = 21.72

attn.o_proj.min_weight = 1.11

attn.o_proj.min_weight_distance = 18.14

mlp.down_proj.max_weight = 1.48

mlp.down_proj.max_weight_position = 21.23

mlp.down_proj.min_weight = 1.47

mlp.down_proj.min_weight_distance = 17.47</p>

</div>

<ul style="margin: 0 0 12px 0; padding-left: 20px;"><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Quantization: BNB_4BIT (during training)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Batch size: 32</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Trials: 230 (final selection: Trial #194)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Datasets: mlabonne/harmless_alpaca + mlabonne/harmful_behaviors</li></ul>

</div>

<ul style="margin: 0 0 12px 0; padding-left: 20px;"><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Type: Qwen3_5ForConditionalGeneration (multimodal with vision encoder)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Parameters: ~9B</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Layers: 32 (hybrid: Gated DeltaNet + Gated Attention)</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Context length: 262,144 tokens</li></ul>

</div>

<th style="padding: 8px 10px; border-bottom: 2px solid #ff6b35; text-align: left; color: #c2410c; font-weight: bold;">Notes</th>

</tr></thead><tbody>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">High quality, near lossless</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">✅ <b>Recommended, best value</b></td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Compact</td>

</tr>

<tr>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">mmproj-BF16</td>

<td style="padding: 7px 10px; border-bottom: 1px solid rgba(128,128,128,0.15); color: #334155; text-align: left;">Vision encoder (multimodal)</td>

</tr>

</tbody></table>

</div>

<h3 style="margin: 16px 0 8px 0; font-size: 14px; color: #1e293b; font-weight: 700;">llama.cpp</h3>

<h3 style="margin: 16px 0 8px 0; font-size: 14px; color: #1e293b; font-weight: 700;">LM Studio</h3>

<p style="margin: 0 0 12px 0; font-size: 13px; color: #334155; line-height: 1.7;">Load the GGUF file directly.</p>

</div>

<ol style="margin: 0 0 12px 0; padding-left: 20px;"><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Safety filters have been significantly reduced via abliteration</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">KL divergence is only 0.0225 — minimal impact on model intelligence</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">The original model supports multimodal (vision); GGUF versions require the mmproj file</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">BenchLocal scores measured at <b>Q6_K</b> on RTX 5070 Ti 16GB with llama.cpp (turboquant). Each scenario was run <b>once with no retries</b> — scores represent single-shot performance</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Outperforms Qwen3.6-35B-A3B (MoE, Thinking ON) and Gemma-4-26B in HermesAgent-20 despite being 1/4 the parameter count</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">TC-11 (Prompt boundary) and TC-12 (Ambiguous request) fixed — areas where the 35B APEX model failed</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Primary weakness: BugFind-15 false positives — the abliterated model tends to over-eagerly "fix" correct code</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;">Please use responsibly</li></ol>

</div>

<ul style="margin: 0 0 12px 0; padding-left: 20px;"><li style="margin-bottom: 4px; font-size: 13px; color: #334155;"><a href="https://github.com/p-e-w/heretic" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">Heretic</a> — Automated censorship removal</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;"><a href="https://huggingface.co/agentscope-ai/QwenPaw-Flash-9B" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">agentscope-ai/QwenPaw-Flash-9B</a> — Base model</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;"><a href="https://github.com/ggml-org/llama.cpp" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">llama.cpp</a> — GGUF quantization and inference</li><li style="margin-bottom: 4px; font-size: 13px; color: #334155;"><a href="https://github.com/stevibe/benchlocal" target="_blank" style="color: #c2410c; text-decoration: none; font-weight: 700;">BenchLocal</a> — Local model agent evaluation suite</li></ul>

</div>

<img

src="https://huggingface.co/SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF/resolve/main/assets/support-sc117.png"

alt="Support SC117 on Ko-fi — support is optional and all releases remain freely available."

width="100%"

</a>

</div>

Run SC117/QwenPaw-Flash-9B-heretic-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models