Quantization publisher

DavidAU

DavidAU publishes 293 quantizations across 36 models in our index, averaging 7.028 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 20 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
39
Quantizations
293
Models covered
36
Avg effective bpw
7.028
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantDavidAUvsTheirsDifference
GLM-4.7-Flash-hereticQ5_129.75 GiBservantofares21.31 GiB+39.6%
GLM-4.7-Flash-hereticQ4_K_M24.06 GiBservantofares17.24 GiB+39.6%
gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-ThinkingQ8_010.53 GiBmradermacher7.48 GiB+40.7%
Qwen3-4B-Apollo-V0.1-4B-Thinking-Heretic-AbliteratedQ8_018.02 GiBHigashikawa17.30 GiB+4.2%
Qwen3-VL-12B-Instruct-Brainstorm20xQ6_K9.75 GiBmradermacher9.07 GiB+7.5%
Qwen3-VL-12B-Instruct-Brainstorm20xQ5_K_S8.36 GiBmradermacher7.68 GiB+8.9%
Qwen3-VL-12B-Instruct-Brainstorm20xQ5_K_M8.54 GiBmradermacher7.86 GiB+8.7%
Qwen3-VL-12B-Instruct-Brainstorm20xQ4_K_S7.09 GiBmradermacher6.40 GiB+10.7%
Qwen3-VL-12B-Instruct-Brainstorm20xQ4_K_M7.41 GiBmradermacher6.72 GiB+10.2%
Qwen3-VL-12B-Instruct-Brainstorm20xIQ4_XS6.76 GiBmradermacher6.12 GiB+10.5%
Qwen3-VL-12B-Instruct-Brainstorm20xQ8_012.28 GiBmradermacher11.74 GiB+4.6%
Qwen3.6-27B-Heretic2-Uncensored-Finetune-ThinkingQ6_K20.86 GiBBugTraceAI20.57 GiB+1.4%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ3_K_S9.76 GiBDevQuasar-99.72 GiB+0.5%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ3_K_M10.83 GiBDevQuasar-910.79 GiB+0.4%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ5_K_S15.53 GiBDevQuasar-915.48 GiB+0.3%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ5_K_M16.00 GiBDevQuasar-915.96 GiB+0.3%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ3_K_L11.73 GiBDevQuasar-911.68 GiB+0.4%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ4_K_M13.65 GiBDevQuasar-913.61 GiB+0.3%
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24BQ4_K_S12.84 GiBDevQuasar-912.80 GiB+0.4%
c4ai-command-r-v01Q2_K12.87 GiBdranger00312.87 GiB-0.0%

A quantization label describes a target, not a recipe. Publishers make different choices about which tensors to keep at higher precision, and some apply an importance matrix while others don't — so two files both honestly labelled the same thing can differ measurably in size and in quality.

Models they publish

ModelQuantizationsSmallest
Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP1022.16 GiB
Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP109.21 GiB
Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking1013.76 GiB
Qwen3.6-27B-Heretic2-Uncensored-Finetune-Thinking109.77 GiB
Llama-3.2-8X3B-MOE-Dark-Champion-Instruct-uncensored-abliterated-18.4B106.56 GiB
GLM-4.7-Flash-heretic149.61 GiB
Qwen3.5-9B-Claude-4.6-Opus-Deckard-V4.2-Uncensored-Heretic-Thinking19.76 GiB
Qwen3.5-9B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING29.77 GiB
Huihui-gpt-oss-20b-BF16-abliterated378.58 GiB
gpt-oss-20b-heretic367.58 GiB
MN-26B-Oblivion-Uncensored109.15 GiB
L3.2-8X4B-MOE-V2-Dark-Champion-Inst-21B-uncen-ablit107.43 GiB
Mistral-MOE-4X7B-Dark-MultiVerse-Uncensored-Enhanced32-24B119.76 GiB
Gemma-The-Writer-N-Restless-Quill-10B-Uncensored104.14 GiB
gemma-4-31B-it-Mystery-Fine-Tune-HERETIC-UNCENSORED-Thinking233.09 GiB
L3.2-Rogue-Creative-Instruct-Uncensored-Abliterated-7B102.73 GiB
Gemma-3-4b-it-Uncensored-DBL-X151.74 GiB
MN-GRAND-23.5B-Gutenberg-UNCENSORED-V2-GLM4.7-Thinking107.87 GiB
c4ai-command-r-v01712.87 GiB
Qwen3-The-Xiaolong-Omega-Directive-22B-uncensored-abliterated117.85 GiB
Llama-3-Stheno-Maid-Blackroot-Grand-HORROR-16B105.88 GiB
Qwen3.5-9B-Star-Trek-TNG-DS9-Heretic-Uncensored-Thinking29.77 GiB
Qwen3-Zero-Coder-Reasoning-0.8B32.33 GiB
Llama-3.2-4X3B-MOE-Hell-California-10B113.53 GiB
gemma-4-E4B-it-The-DECKARD-Expresso-Universe-HERETIC-UNCENSORED-Thinking110.53 GiB
L3-V2-Stheno-Maid-Blackroot-Grand-HORROR-17B-GLM4.7-Thinking-UNCENSORED105.60 GiB
Qwen3-4B-Apollo-V0.1-4B-Thinking-Heretic-Abliterated217.30 GiB
Qwen2.5-MOE-2X1.5B-DeepSeek-Uncensored-Censored-4B121.48 GiB
Qwen3-48B-A4B-Savant-Commander-Distill-12X-Closed-Open-Heretic-Uncensored815.06 GiB
L3-Dark-Planet-8B113.60 GiB