← back to catalog · registered 2026-11-20 03:04

Alex500/Qwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-GGUF

Alex500 Qwen 9B GGUF second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Alex500%2FQwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-GGUF"
Response includes
  • classification m3
  • files 6
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
988
Likes
1
Model age
2d ago
created 2026-11-20
Downloads over time
Now 86 → from 0 ↑ 0%
0326395 0 on Nov 20 86 on Nov 21 Nov
Nov 20 → Nov 21 · 2 snapshots · spans 1 day

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Metadata

License
apache-2.0
Languages
en pt
Quantizations
Q5_K Q6_K
Tags
gguf qwen3 coding rust function-calling tool-use agents mcp bionic-devtools bionic-bridge reasoning cot

Related

Total size
12.9 GB
Files
6
Quantizations
3
Registered
2026-11-20 03:04
Last updated on HF
2026-11-21 04:45

Files by quantization

Q6_K 1 file 6.85 GB
Qwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-Q6_K.gguf 6.85 GB 2527b973 download
Q5_K 1 file 6.02 GB
Qwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-Q5_K_M.gguf 6.02 GB ab2872ab download
Auxiliary files 4 files 4.73 MB
evaluation_dataset.jsonl 4.63 MB fcb56a9e download
benchmark_results.json 97.0 KB 9c140d1a download
README.md 7.29 KB 01ee344a download
.gitattributes 1.67 KB 05db618c download

README current version from Hugging Face


license: apache-2.0
base_model: benjamin920101/Qwen3.8-9B-Distill-uncensored-heretic-china_uncensored_dpo
tags:

  • qwen3
  • gguf
  • coding
  • rust
  • function-calling
  • tool-use
  • agents
  • mcp
  • reasoning
  • cot
  • uncensored
    pipeline_tag: text-generation
    language:
  • en
  • pt

Qwen 3.8 9B Uncensored Heretic Ch. Enhanced V1 (GGUF)

📌 Executive Summary

🇧🇷 Resumo em Português

O Qwen 3.8 9B Uncensored Heretic Ch. Enhanced V1 é uma versão especializada e aprimorada via fine-tuning QLoRA de alto rigor a partir do modelo base Qwen3.8-9B-Distill-uncensored-heretic-china_uncensored_dpo. Esta versão foi intensamente treinada em 5 domínios críticos: Programação avançada (Rust First, Python, TypeScript e SQL), Lógica algorítmica e raciocínio profundo (Chain-of-Thought), Chamada de ferramentas e schemas estritos (Tool-Use / Function Calling), Orquestração autônoma de subagentes e Protocolo MCP (Model Context Protocol) & Skills. O modelo mantém sua natureza sem censura prévia, ao mesmo tempo em que eleva a confiabilidade de código e conformidade sintática, demonstrando um salto global de +137% de evolução matemática em testes de benchmark.

🇺🇸 English Summary

Qwen 3.8 9B Uncensored Heretic Ch. Enhanced V1 is a specialized, post-trained iteration of Qwen3.8-9B-Distill-uncensored-heretic-china_uncensored_dpo, enhanced through precision QLoRA fine-tuning. It has been trained across five core technical pillars: Advanced Programming (Idiomatic Rust First, Python, TypeScript, SQL), Algorithmic Logic and Deep Chain-of-Thought reasoning, Structured Function Calling and Tool-Use, Autonomous Multi-Agent Orchestration, and the Model Context Protocol (MCP) & Agentic Skills. The model preserves its uncensored foundation while dramatically improving structured reasoning, tool syntax compliance, and coding precision, achieving a +137% measured global intelligence increase across comprehensive benchmark evaluations.


📊 Benchmark Evaluations & Quantitative Intelligence Gains

The model was subjected to a rigorous 50-scenario benchmark suite covering the 5 targeted specialization areas, evaluated under identical criteria before and after training.

Comparative Performance Matrix

Evaluation Category Baseline (Pre-Training) Enhanced V1 (Post-Training) Absolute Gain Relative Gain (%)
🔌 Model Context Protocol (MCP) & Skills 0.20 / 10.0 2.65 / 10.0 +2.45 pts +1225.0% 🚀
🤖 Autonomous Multi-Agent Orchestration 0.30 / 10.0 2.67 / 10.0 +2.37 pts +790.0% 🚀
🛠️ Tool-Use & JSON Function Calling 1.20 / 10.0 5.08 / 10.0 +3.88 pts +323.3% 🚀
🦀 Programming (Rust / Python / TS / SQL) 3.80 / 10.0 6.26 / 10.0 +2.46 pts +64.7% 🚀
🧩 Algorithmic Logic & CoT Reasoning 4.10 / 10.0 6.10 / 10.0 +2.00 pts +48.8% 🚀
🏆 GLOBAL COMPOSITE AVERAGE 1.92 / 10.0 4.55 / 10.0 +2.63 pts +137.0% Overall Gain 🚀

Training Loss & Token Metrics

  • Initial Training Loss: 1.205
  • Final Training Loss: 0.6185 (-51.3% error reduction)
  • Validation Loss: 0.5857 (confirmed generalization without overfitting)
  • Token Prediction Accuracy: Jumped from 68.97% to 83.51% (+14.54 percentage points)
  • Total Processed Tokens: 4,290,000+ tokens across 1,000 optimization steps

🧠 Training Methodology & Architecture

Base Architecture

  • Base Model: Qwen 3.8 9B Distill Uncensored Heretic DPO
  • Architecture: Hybrid SSM (Gated DeltaNet / Mamba-style recurrent layers) + Full Attention Transformer layers
  • Total Parameters: 9.4 Billion

QLoRA Fine-Tuning Setup

  • Quantization: NF4 (4-bit NormalFloat) with double quantization and torch.bfloat16 compute dtype
  • Adapter Targets: All projection layers (q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj)
  • LoRA Hyperparameters:
    • Rank ($r$): 16
    • Alpha ($\alpha$): 32
    • Dropout: 0.05
    • Trainable Parameters: 29,622,272 (~0.31% of total parameters)
  • Optimization:
    • Optimizer: adamw_torch (PyTorch native)
    • Learning Rate: 2e-5 with cosine decay and warmup
    • Batching: Micro-batch size 1, Gradient Accumulation steps 8 (Effective Batch Size = 8)
    • Sequence Length: 896 tokens
    • Precision: Native bfloat16 mixed precision

Fusion & Quantization Pipeline

  1. Weight Merging: The LoRA adapter was dequantized and merged in full 16-bit precision (bfloat16) with the base model weights, producing a unified 17.9 GB model.
  2. GGUF Native Quantization: Converted via llama.cpp using exact 427-tensor layout compatibility:
    • Q6_K (~6.85 GB): High-precision quantization for maximum adherence to strict coding standards and JSON schemas.
    • Q5_K_M (~6.02 GB): Compact, fast quantization optimized for quick interactive inference.

📚 Dataset Curation & Processing

A curated dataset of 18,491 total high-signal instruction-response pairs was created and normalized into canonical ChatML format:

  1. Coding Domain (code_feedback) — 5,991 pairs:
    • Idiomatic Rust implementations (memory safety, zero-cost abstractions, Axum, Serde, Tokio async, concurrency).
    • Python debugging, optimization, algorithm development, and SQL database schema optimization.
  2. Function Calling & Tool-Use (xlam_tool_use) — 4,000 pairs:
    • Multi-parameter tool calls, schema validation, structured error recovery, and strict JSON output formats.
  3. Chain-of-Thought Reasoning (reasoning_data) — 3,500 pairs:
    • Step-by-step mathematical proofs, graph theory, time/space complexity derivations, and distributed systems logic within <think>...</think> traces.
  4. Synthetic Agents & MCP Protocols (synthetic_agents_mcp) — 3,000 pairs:
    • Autonomous agent decomposition (DAG workflows, orchestrator-worker patterns, judge agents).
    • Model Context Protocol (MCP) server-client communications, tool listings, schema mapping, and resource handlers.
  5. Anti-Catastrophic Forgetting (general_finetome) — 2,000 pairs:
    • General knowledge and multi-turn conversational data to ensure fluent dialogue and preserve general intelligence.

Data Split:

  • Training Set: 17,566 examples (datasets/processed/train.jsonl)
  • Evaluation Set: 925 examples (datasets/processed/eval.jsonl)

📦 Provided Files

Filename Quantization Size Description
Qwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-Q6_K.gguf Q6_K ~6.85 GB Recommended: Maximum fidelity for code generation, function calling & MCP.
Qwen-3.8-9B-Uncensored-Heretic-Ch-Enhanced-V1-Q5_K_M.gguf Q5_K_M ~6.02 GB Fast & efficient balance for general chat and local agent execution.

🚀 How to Run in LM Studio / llama.cpp

Recommended Inference Parameters

  • Prompt Format: ChatML (<|im_start|>system...<|im_end|><|im_start|>user...<|im_end|><|im_start|>assistant...)
  • Temperature: 0.3 to 0.5 (for coding and structured tool-use) | 0.7 (for creative writing)
  • Top P: 0.9
  • Context Length: 8192 to 16384 tokens
  • GPU Offload: 100% offloadable on any 12GB+ GPU (e.g., RTX 3060 12GB, RTX 4070, RTX 4080).

README history 11 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-11-21 Upload README.md with huggingface_hub e861273 3.9 KB
    Loading...
  2. 2026-11-20 Upload README.md with huggingface_hub a3bc721 6.7 KB
    Loading...
  3. 2026-11-20 Upload README.md with huggingface_hub 1e54dcf 7.2 KB
    Loading...
  4. 2026-11-20 Upload README.md with huggingface_hub 00d8990 6.9 KB
    Loading...
  5. 2026-11-20 Upload README.md with huggingface_hub 9d08c9d 2.5 KB
    Loading...
  6. 2026-11-20 Upload README.md with huggingface_hub 95d9fdd 4.4 KB
    Loading...
  7. 2026-11-20 Update Model Card for V4 Surgical Refinement (Zero-Hesitation, Anti-Refusal, ... 57de632 9.1 KB
    Loading...
  8. 2026-11-20 Update Model Card for V3 Surgical Refinement (5.01 global benchmark, zero-sho... 34c6aa6 10.5 KB
    Loading...
  9. 2026-11-20 Feature: Add Whats New / Novidades section to README 92935c2 11.6 KB
    Loading...
  10. 2026-11-20 Update Model Card with Surgical Refinement V2, 600-step benchmarks, and crash... d52d2d9 9 KB
    Loading...
  11. 2026-11-20 Add comprehensive Model Card with PT/EN summaries and benchmarks 338bf0b 7.3 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration