GraySoft
Projects Models About FAQ Contact Download guIDE →

llmfan46/qwen3.5-27b-writer-v2-uncensored-heretic-gguf Q3_K_L GGUF - Free GGUF Download is indexed on GraySoft with repository links, GGUF quant files, and Hugging Face metadata. This page helps you pick a local model for guIDE or other runtimes. See related models in the same shard below.

Model Intelligence Sheet

llmfan46/qwen3.5-27b-writer-v2-uncensored-heretic-gguf overview

Comprehensive model page for llmfan46/qwen3.5-27b-writer-v2-uncensored-heretic-gguf

ggufqwen3_5hereticuncensoreddecensoredabliteratedaradataset:ConicCat/Gutenberg-SFTdataset:ConicCat/AntiRepdataset:ConicCat/Condor-SFT-Filtereddataset:ConicCat/MiniC2_V3.2base_model:llmfan46/Qwen3.5-27B-Writer-V2-uncensored-hereticbase_model:quantized:llmfan46/Qwen3.5-27B-Writer-V2-uncensored-hereticlicense:apache-2.0endpoints_compatibleregion:usconversational
llmfan46/qwen3.5-27b-writer-v2-uncensored-heretic-gguf visual
Downloads
818
Likes
1
Pipeline
Library
Visibility
Public
Access
Open

Repository Files & Downloads

10 files detected
Direct downloads for all repository files
FileTypeQuantizationSizeLink
Qwen3.5-27B-Writer-V2-mmproj-BF16.gguf GGUF BF16 888.01 MB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-BF16.gguf GGUF BF16 50.11 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q3_K_L.gguf GGUF Q3_K_L 13.36 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q3_K_M.gguf GGUF Q3_K_M 12.39 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q4_K_M.gguf GGUF Q4_K_M 15.41 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q4_K_S.gguf GGUF Q4_K_S 14.52 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q5_K_M.gguf GGUF Q5_K_M 17.91 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q5_K_S.gguf GGUF Q5_K_S 17.40 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q6_K.gguf GGUF Q6_K 20.57 GB Download
Qwen3.5-27B-Writer-V2-uncensored-heretic-Q8_0.gguf GGUF 26.63 GB Download

Model Details Live

Model Slug
llmfan46/qwen3.5-27b-writer-v2-uncensored-heretic-gguf
Author
llmfan46
Pipeline Task
Library
Created
2026-04-16
Last Modified
2026-04-16
Gated
No
Private
No
HF SHA
2a299bb2101809987146a96ad54b60f4571b6c2f
License
apache-2.0
Language
Unknown
Base Model
llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic

Metadata Inspector

Normalized metadata (stored in metadata_json)
{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "datasets": [
      "ConicCat/Gutenberg-SFT",
      "ConicCat/AntiRep",
      "ConicCat/Condor-SFT-Filtered",
      "ConicCat/MiniC2_V3.2"
    ],
    "base_model": [
      "llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic"
    ],
    "tags": [
      "qwen3_5",
      "heretic",
      "uncensored",
      "decensored",
      "abliterated",
      "ara"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "datasets": [
        "ConicCat/Gutenberg-SFT",
        "ConicCat/AntiRep",
        "ConicCat/Condor-SFT-Filtered",
        "ConicCat/MiniC2_V3.2"
      ],
      "base_model": [
        "llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic"
      ],
      "tags": [
        "qwen3_5",
        "heretic",
        "uncensored",
        "decensored",
        "abliterated",
        "ara"
      ]
    },
    "hero_image_url": "https://huggingface.co/llmfan46/Omega-Darker-Gaslight_The-Final-Forgotten-Fever-Dream-24B-ultra-uncensored-heretic-v1/resolve/main/waifu001.webp",
    "summary": "",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\ndatasets:\n- ConicCat/Gutenberg-SFT\n- ConicCat/AntiRep\n- ConicCat/Condor-SFT-Filtered\n- ConicCat/MiniC2_V3.2\nbase_model:\n- llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic\ntags:\n- qwen3_5\n- heretic\n- uncensored\n- decensored\n- abliterated\n- ara\n---\n<div style=\"background-color: #ff4444; color: white; padding: 20px; border-radius: 10px; text-align: center; margin: 20px 0;\">\n<h2 style=\"color: white; margin: 0 0 10px 0;\">🚨⚠️ I HAVE REACHED HUGGING FACE'S FREE STORAGE LIMIT ⚠️🚨</h2>\n<p style=\"font-size: 18px; margin: 0 0 15px 0;\">I can no longer upload new models unless I can cover the cost of additional storage.<br>I host <b>70+ free models</b> as an independent contributor and this work is unpaid.<br><b>Without your support, no more new models can be uploaded.</b></p>\n<p style=\"font-size: 20px; margin: 0;\">\n<a href=\"https://patreon.com/LLMfan46\" style=\"color: white; text-decoration: underline;\">🎉 Patreon (Monthly)</a> &nbsp;|&nbsp;\n<a href=\"https://ko-fi.com/llmfan46\" style=\"color: white; text-decoration: underline;\">☕ Ko-fi (One-time)</a>\n</p>\n<p style=\"font-size: 16px; margin: 10px 0 0 0;\">Every contribution goes directly toward Hugging Face storage fees to keep models free for everyone.</p>\n</div>\n\n---\n\n### **91% fewer refusals** (8/100 Uncensored vs 93/100 Original) while preserving model quality (0.0274 KL divergence).\n\n## ❤️ Support My Work\nCreating these models takes significant time, work and compute. If you find them useful consider supporting me:\n\n![image/png](https://huggingface.co/llmfan46/Omega-Darker-Gaslight_The-Final-Forgotten-Fever-Dream-24B-ultra-uncensored-heretic-v1/resolve/main/waifu001.webp)\n\n| Platform | Link | What you get |\n|----------|------|--------------|\n| 🎉 Patreon | [Monthly support](https://patreon.com/LLMfan46) | Priority model requests |\n| ☕ Ko-fi | [One-time tip](https://ko-fi.com/llmfan46) | My eternal gratitude |\n\nYour help will motivate me and would go into further improving my workflow and coverings fees for storage, compute and may even help uncensoring bigger model with rental Cloud GPUs.\n\n-----\n\nGGUF quantizations of [llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic](https://huggingface.co/llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic).\n\nThis model is great for creative writing and translation, the original base model writing and translations feels a litle stiff which might not really read very nicely some times, Qwen3.5-27B-Writer-V2-uncensored-heretic aims to fix this issue and improve the writing quality of Qwen3.5-27B.\n\n# This is a decensored version of [ConicCat/Qwen3.5-27B-Writer-V2](https://huggingface.co/ConicCat/Qwen3.5-27B-Writer-V2), made using [Heretic](https://github.com/p-e-w/heretic) v1.2.0 with the [Arbitrary-Rank Ablation (ARA)](https://github.com/p-e-w/heretic/pull/211) method\n\n## Abliteration parameters\n\n| Parameter | Value |\n| :-------- | :---: |\n| **start_layer_index** | 31 |\n| **end_layer_index** | 56 |\n| **preserve_good_behavior_weight** | 0.4059 |\n| **steer_bad_behavior_weight** | 0.0001 |\n| **overcorrect_relative_weight** | 1.1869 |\n| **neighbor_count** | 10 |\n\n## Targeted components\n\n  * attn.o_proj\n  * attn.out_proj\n\n## Performance\n\n| Metric | This model | Original model ([ConicCat/Qwen3.5-27B-Writer-V2](https://huggingface.co/ConicCat/Qwen3.5-27B-Writer-V2)) |\n| :----- | :--------: | :---------------------------: |\n| **KL divergence** | <span style=\"color:darkgoldenrod\">0.0274</span> | 0 *(by definition)* |\n| **Refusals** | ✅ <span style=\"color:darkgreen\">8/100</span> | ❌ <span style=\"color:blue\">93/100</span> |\n\nLower refusals indicate fewer content restrictions, while lower KL divergence indicates more closeness to the original model's baseline. Higher refusals cause more rejections, objections, pushbacks, lecturing, censorship, softening and deflections.\n\n## MMLU test results:\n\n<span style=\"color:blue\">Original:</span>\n\n|                 Tasks                 |Version|Filter|n-shot|Metric|   |Value |   |Stderr|\n|---------------------------------------|------:|------|-----:|------|---|-----:|---|-----:|\n|mmlu                                   |      2|none  |      |acc   |↑  |0.8562|±  |0.0028|\n| - humanities                          |      2|none  |      |acc   |↑  |0.8047|±  |0.0056|\n|  - formal_logic                       |      1|none  |     0|acc   |↑  |0.7302|±  |0.0397|\n|  - high_school_european_history       |      1|none  |     0|acc   |↑  |0.9030|±  |0.0231|\n|  - high_school_us_history             |      1|none  |     0|acc   |↑  |0.9412|±  |0.0165|\n|  - high_school_world_history          |      1|none  |     0|acc   |↑  |0.9409|±  |0.0153|\n|  - international_law                  |      1|none  |     0|acc   |↑  |0.9256|±  |0.0240|\n|  - jurisprudence                      |      1|none  |     0|acc   |↑  |0.9074|±  |0.0280|\n|  - logical_fallacies                  |      1|none  |     0|acc   |↑  |0.9202|±  |0.0213|\n|  - moral_disputes                     |      1|none  |     0|acc   |↑  |0.8584|±  |0.0188|\n|  - moral_scenarios                    |      1|none  |     0|acc   |↑  |0.7352|±  |0.0148|\n|  - philosophy                         |      1|none  |     0|acc   |↑  |0.8842|±  |0.0182|\n|  - prehistory                         |      1|none  |     0|acc   |↑  |0.9167|±  |0.0154|\n|  - professional_law                   |      1|none  |     0|acc   |↑  |0.7080|±  |0.0116|\n|  - world_religions                    |      1|none  |     0|acc   |↑  |0.9181|±  |0.0210|\n| - other                               |      2|none  |      |acc   |↑  |0.8735|±  |0.0057|\n|  - business_ethics                    |      1|none  |     0|acc   |↑  |0.8300|±  |0.0378|\n|  - clinical_knowledge                 |      1|none  |     0|acc   |↑  |0.8868|±  |0.0195|\n|  - college_medicine                   |      1|none  |     0|acc   |↑  |0.8382|±  |0.0281|\n|  - global_facts                       |      1|none  |     0|acc   |↑  |0.6200|±  |0.0488|\n|  - human_aging                        |      1|none  |     0|acc   |↑  |0.8430|±  |0.0244|\n|  - management                         |      1|none  |     0|acc   |↑  |0.8738|±  |0.0329|\n|  - marketing                          |      1|none  |     0|acc   |↑  |0.9530|±  |0.0139|\n|  - medical_genetics                   |      1|none  |     0|acc   |↑  |0.9700|±  |0.0171|\n|  - miscellaneous                      |      1|none  |     0|acc   |↑  |0.9387|±  |0.0086|\n|  - nutrition                          |      1|none  |     0|acc   |↑  |0.9020|±  |0.0170|\n|  - professional_accounting            |      1|none  |     0|acc   |↑  |0.8014|±  |0.0238|\n|  - professional_medicine              |      1|none  |     0|acc   |↑  |0.9522|±  |0.0130|\n|  - virology                           |      1|none  |     0|acc   |↑  |0.5723|±  |0.0385|\n| - social sciences                     |      2|none  |      |acc   |↑  |0.9162|±  |0.0049|\n|  - econometrics                       |      1|none  |     0|acc   |↑  |0.8158|±  |0.0365|\n|  - high_school_geography              |      1|none  |     0|acc   |↑  |0.9596|±  |0.0140|\n|  - high_school_government_and_politics|      1|none  |     0|acc   |↑  |0.9896|±  |0.0073|\n|  - high_school_macroeconomics         |      1|none  |     0|acc   |↑  |0.9282|±  |0.0131|\n|  - high_school_microeconomics         |      1|none  |     0|acc   |↑  |0.9664|±  |0.0117|\n|  - high_school_psychology             |      1|none  |     0|acc   |↑  |0.9541|±  |0.0090|\n|  - human_sexuality                    |      1|none  |     0|acc   |↑  |0.9160|±  |0.0243|\n|  - professional_psychology            |      1|none  |     0|acc   |↑  |0.8725|±  |0.0135|\n|  - public_relations                   |      1|none  |     0|acc   |↑  |0.7636|±  |0.0407|\n|  - security_studies                   |      1|none  |     0|acc   |↑  |0.8449|±  |0.0232|\n|  - sociology                          |      1|none  |     0|acc   |↑  |0.9652|±  |0.0130|\n|  - us_foreign_policy                  |      1|none  |     0|acc   |↑  |0.9400|±  |0.0239|\n| - stem                                |      2|none  |      |acc   |↑  |0.8576|±  |0.0060|\n|  - abstract_algebra                   |      1|none  |     0|acc   |↑  |0.8000|±  |0.0402|\n|  - anatomy                            |      1|none  |     0|acc   |↑  |0.8296|±  |0.0325|\n|  - astronomy                          |      1|none  |     0|acc   |↑  |0.9671|±  |0.0145|\n|  - college_biology                    |      1|none  |     0|acc   |↑  |0.9792|±  |0.0119|\n|  - college_chemistry                  |      1|none  |     0|acc   |↑  |0.6800|±  |0.0469|\n|  - college_computer_science           |      1|none  |     0|acc   |↑  |0.8300|±  |0.0378|\n|  - college_mathematics                |      1|none  |     0|acc   |↑  |0.6800|±  |0.0469|\n|  - college_physics                    |      1|none  |     0|acc   |↑  |0.8235|±  |0.0379|\n|  - computer_security                  |      1|none  |     0|acc   |↑  |0.8700|±  |0.0338|\n|  - conceptual_physics                 |      1|none  |     0|acc   |↑  |0.9404|±  |0.0155|\n|  - electrical_engineering             |      1|none  |     0|acc   |↑  |0.8276|±  |0.0315|\n|  - elementary_mathematics             |      1|none  |     0|acc   |↑  |0.9101|±  |0.0147|\n|  - high_school_biology                |      1|none  |     0|acc   |↑  |0.9516|±  |0.0122|\n|  - high_school_chemistry              |      1|none  |     0|acc   |↑  |0.8522|±  |0.0250|\n|  - high_school_computer_science       |      1|none  |     0|acc   |↑  |0.9300|±  |0.0256|\n|  - high_school_mathematics            |      1|none  |     0|acc   |↑  |0.6741|±  |0.0286|\n|  - high_school_physics                |      1|none  |     0|acc   |↑  |0.8609|±  |0.0283|\n|  - high_school_statistics             |      1|none  |     0|acc   |↑  |0.8704|±  |0.0229|\n|  - machine_learning                   |      1|none  |     0|acc   |↑  |0.7857|±  |0.0389|\n\n|      Groups      |Version|Filter|n-shot|Metric|   |Value |   |Stderr|\n|------------------|------:|------|------|------|---|-----:|---|-----:|\n|mmlu              |      2|none  |      |acc   |↑  |0.8562|±  |0.0028|\n| - humanities     |      2|none  |      |acc   |↑  |0.8047|±  |0.0056|\n| - other          |      2|none  |      |acc   |↑  |0.8735|±  |0.0057|\n| - social sciences|      2|none  |      |acc   |↑  |0.9162|±  |0.0049|\n| - stem           |      2|none  |      |acc   |↑  |0.8576|±  |0.0060|\n\n\n<span style=\"color:darkgreen\">Heretic:</span>\n\n|                 Tasks                 |Version|Filter|n-shot|Metric|   |Value |   |Stderr|\n|---------------------------------------|------:|------|-----:|------|---|-----:|---|-----:|\n|mmlu                                   |      2|none  |      |acc   |↑  |0.8469|±  |0.0029|\n| - humanities                          |      2|none  |      |acc   |↑  |0.7858|±  |0.0058|\n|  - formal_logic                       |      1|none  |     0|acc   |↑  |0.7302|±  |0.0397|\n|  - high_school_european_history       |      1|none  |     0|acc   |↑  |0.8970|±  |0.0237|\n|  - high_school_us_history             |      1|none  |     0|acc   |↑  |0.9412|±  |0.0165|\n|  - high_school_world_history          |      1|none  |     0|acc   |↑  |0.9367|±  |0.0158|\n|  - international_law                  |      1|none  |     0|acc   |↑  |0.9256|±  |0.0240|\n|  - jurisprudence                      |      1|none  |     0|acc   |↑  |0.9167|±  |0.0267|\n|  - logical_fallacies                  |      1|none  |     0|acc   |↑  |0.8957|±  |0.0240|\n|  - moral_disputes                     |      1|none  |     0|acc   |↑  |0.8526|±  |0.0191|\n|  - moral_scenarios                    |      1|none  |     0|acc   |↑  |0.6458|±  |0.0160|\n|  - philosophy                         |      1|none  |     0|acc   |↑  |0.8810|±  |0.0184|\n|  - prehistory                         |      1|none  |     0|acc   |↑  |0.9043|±  |0.0164|\n|  - professional_law                   |      1|none  |     0|acc   |↑  |0.7086|±  |0.0116|\n|  - world_religions                    |      1|none  |     0|acc   |↑  |0.9298|±  |0.0196|\n| - other                               |      2|none  |      |acc   |↑  |0.8725|±  |0.0057|\n|  - business_ethics                    |      1|none  |     0|acc   |↑  |0.8200|±  |0.0386|\n|  - clinical_knowledge                 |      1|none  |     0|acc   |↑  |0.9057|±  |0.0180|\n|  - college_medicine                   |      1|none  |     0|acc   |↑  |0.8613|±  |0.0264|\n|  - global_facts                       |      1|none  |     0|acc   |↑  |0.5600|±  |0.0499|\n|  - human_aging                        |      1|none  |     0|acc   |↑  |0.8341|±  |0.0250|\n|  - management                         |      1|none  |     0|acc   |↑  |0.9223|±  |0.0265|\n|  - marketing                          |      1|none  |     0|acc   |↑  |0.9573|±  |0.0133|\n|  - medical_genetics                   |      1|none  |     0|acc   |↑  |0.9700|±  |0.0171|\n|  - miscellaneous                      |      1|none  |     0|acc   |↑  |0.9425|±  |0.0083|\n|  - nutrition                          |      1|none  |     0|acc   |↑  |0.9020|±  |0.0170|\n|  - professional_accounting            |      1|none  |     0|acc   |↑  |0.7766|±  |0.0248|\n|  - professional_medicine              |      1|none  |     0|acc   |↑  |0.9338|±  |0.0151|\n|  - virology                           |      1|none  |     0|acc   |↑  |0.5723|±  |0.0385|\n| - social sciences                     |      2|none  |      |acc   |↑  |0.9110|±  |0.0050|\n|  - econometrics                       |      1|none  |     0|acc   |↑  |0.8070|±  |0.0371|\n|  - high_school_geography              |      1|none  |     0|acc   |↑  |0.9495|±  |0.0156|\n|  - high_school_government_and_politics|      1|none  |     0|acc   |↑  |0.9845|±  |0.0089|\n|  - high_school_macroeconomics         |      1|none  |     0|acc   |↑  |0.9205|±  |0.0137|\n|  - high_school_microeconomics         |      1|none  |     0|acc   |↑  |0.9664|±  |0.0117|\n|  - high_school_psychology             |      1|none  |     0|acc   |↑  |0.9486|±  |0.0095|\n|  - human_sexuality                    |      1|none  |     0|acc   |↑  |0.9084|±  |0.0253|\n|  - professional_psychology            |      1|none  |     0|acc   |↑  |0.8742|±  |0.0134|\n|  - public_relations                   |      1|none  |     0|acc   |↑  |0.7727|±  |0.0401|\n|  - security_studies                   |      1|none  |     0|acc   |↑  |0.8204|±  |0.0246|\n|  - sociology                          |      1|none  |     0|acc   |↑  |0.9602|±  |0.0138|\n|  - us_foreign_policy                  |      1|none  |     0|acc   |↑  |0.9400|±  |0.0239|\n| - stem                                |      2|none  |      |acc   |↑  |0.8503|±  |0.0061|\n|  - abstract_algebra                   |      1|none  |     0|acc   |↑  |0.7100|±  |0.0456|\n|  - anatomy                            |      1|none  |     0|acc   |↑  |0.8444|±  |0.0313|\n|  - astronomy                          |      1|none  |     0|acc   |↑  |0.9605|±  |0.0158|\n|  - college_biology                    |      1|none  |     0|acc   |↑  |0.9722|±  |0.0137|\n|  - college_chemistry                  |      1|none  |     0|acc   |↑  |0.6400|±  |0.0482|\n|  - college_computer_science           |      1|none  |     0|acc   |↑  |0.8300|±  |0.0378|\n|  - college_mathematics                |      1|none  |     0|acc   |↑  |0.7100|±  |0.0456|\n|  - college_physics                    |      1|none  |     0|acc   |↑  |0.8529|±  |0.0352|\n|  - computer_security                  |      1|none  |     0|acc   |↑  |0.8600|±  |0.0349|\n|  - conceptual_physics                 |      1|none  |     0|acc   |↑  |0.9362|±  |0.0160|\n|  - electrical_engineering             |      1|none  |     0|acc   |↑  |0.8276|±  |0.0315|\n|  - elementary_mathematics             |      1|none  |     0|acc   |↑  |0.9074|±  |0.0149|\n|  - high_school_biology                |      1|none  |     0|acc   |↑  |0.9387|±  |0.0136|\n|  - high_school_chemistry              |      1|none  |     0|acc   |↑  |0.8473|±  |0.0253|\n|  - high_school_computer_science       |      1|none  |     0|acc   |↑  |0.9200|±  |0.0273|\n|  - high_school_mathematics            |      1|none  |     0|acc   |↑  |0.6630|±  |0.0288|\n|  - high_school_physics                |      1|none  |     0|acc   |↑  |0.8411|±  |0.0299|\n|  - high_school_statistics             |      1|none  |     0|acc   |↑  |0.8704|±  |0.0229|\n|  - machine_learning                   |      1|none  |     0|acc   |↑  |0.7768|±  |0.0395|\n\n|      Groups      |Version|Filter|n-shot|Metric|   |Value |   |Stderr|\n|------------------|------:|------|------|------|---|-----:|---|-----:|\n|mmlu              |      2|none  |      |acc   |↑  |0.8469|±  |0.0029|\n| - humanities     |      2|none  |      |acc   |↑  |0.7858|±  |0.0058|\n| - other          |      2|none  |      |acc   |↑  |0.8725|±  |0.0057|\n| - social sciences|      2|none  |      |acc   |↑  |0.9110|±  |0.0050|\n| - stem           |      2|none  |      |acc   |↑  |0.8503|±  |0.0061|\n\nMMLU - Massive Multitask Language Understanding, multiple-choice questions across 57 subjects (math, history, law, medicine, etc.).\n\n-----\n\n## Quantizations\n\n| Filename | Quant | Description |\n|----------|-------|-------------|\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-BF16.gguf | BF16 | Full precision |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q8_0.gguf | Q8_0 | Near-lossless, recommended |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q6_K.gguf | Q6_K | Excellent quality |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q5_K_M.gguf | Q5_K_M | Good balance |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q5_K_SQwen3.5-27B-ultra-uncensored-heretic-v2-v2-Q5_K_S.gguf | Q5_K_S | Smaller Q5 |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q4_K_M.gguf | Q4_K_M | Good for limited VRAM |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q4_K_S.gguf | Q4_K_S | Smaller Q4 |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q3_K_L.gguf | Q3_K_L | Low VRAM, decent quality |\n| Qwen3.5-27B-Writer-V2-uncensored-heretic-Q3_K_M.gguf | Q3_K_M | Low VRAM, smaller |\n\n## Vision Projector\n\n| Filename | Quant | Description |\n|----------|-------|-------------|\n| Qwen3.5-27B-Writer-V2-mmproj-BF16.gguf | BF16 | Native precision |\n\nA Vision Projector File is Required for vision/multimodal capabilities. Use alongside any quantization above.\n\n## Usage\n\nWorks with llama.cpp, LM Studio, Ollama, and other GGUF-compatible tools.\n\n-----\n\n# ConicCat/Qwen3.5-27B-Writer-V2\n\nA tentative second version. Hopefully, it's better.\n\nA writing & roleplay finetune of Qwen3.5 27B. The primary emphasis is on writing quality as it strongly generalizes across both domains.\n\nThe basic idea is to use a curriculum learning setup to overcome the lack of high quality roleplay data by first training on lower quality\nroleplay data, then training on higher quality writing data. Starting from ConicCat/Qwen3.5-Antirep-27B, the model was trained on a roughly equal mixture of instruct / roleplay / writing data for three epochs. The model was then trained for \neleven epochs on a smaller dataset of book chunks.\n\n### Recommended Settings\n\n* Chatml template with `<think>\\n\\n</think>\\n` prefill or `<think>\\n` prefill. Should think less!\n* temperature = `0.7`\n* top_p = `0.95`\n* A moderate dry penalty of ~ `0.4-0.8` should work well.\n* For quants, Q4_K_M runs well with `~100k` context on 24GB Vram\n* IQ4_XS should fit on 16GB Vram with about `20-24k` context with the vulkan backend, although it's pretty tight and may require some fiddling around with open programs e.t.c.\n\n### Datasets\n\n* ConicCat/AntiRep to mitigate repetitition.\n* internlm/Condor-SFT-20K for instruct; even though instruct capabilities are not the primary focus, adding some instruct data helps mitigate forgetting and maintains general intellect and instruction following capabilites.\n* ConicCat/Gutenberg-SFT. A reformatted version of the original Gutenberg DPO dataset by jondurbin for SFT with some slight augmentation to address many of the samples being overly long.\n* ConicCat/MiniC2_V3.2. The venerable C2, with cleaned and reformatted system prompts, and all user / assistant turns replaced by V3.2.\n\n* A dataset of backtranslated books. Unfortunately, I am unable to release this set as all of the data is under copyright.",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "qwen3_5",
    "heretic",
    "uncensored",
    "decensored",
    "abliterated",
    "ara",
    "dataset:ConicCat/Gutenberg-SFT",
    "dataset:ConicCat/AntiRep",
    "dataset:ConicCat/Condor-SFT-Filtered",
    "dataset:ConicCat/MiniC2_V3.2",
    "base_model:llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic",
    "base_model:quantized:llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 1,
  "downloads": 818,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-16T11:41:09.000Z",
  "created_at": "2026-04-16T11:25:01.000Z",
  "pipeline_tag": "",
  "library_name": ""
}
Source payload excerpt (from Hugging Face API)
{
  "_id": "69e0c70df6419d1d7c0d1b2a",
  "id": "llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic-GGUF",
  "modelId": "llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic-GGUF",
  "sha": "2a299bb2101809987146a96ad54b60f4571b6c2f",
  "createdAt": "2026-04-16T11:25:01.000Z",
  "lastModified": "2026-04-16T11:41:09.000Z",
  "author": "llmfan46",
  "downloads": 818,
  "likes": 1,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 12
}