Quantization publisher
mudler
mudler publishes 52 quantizations across 16 models in our index, averaging 13.619 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 6 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
19
Quantizations
52
Models covered
16
Avg effective bpw
13.619
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | mudler | vs | Theirs | Difference |
|---|---|---|---|---|---|
| VibeVoice-ASR | Q8_0 | 14.55 GiB | cstr | 8.80 GiB | +65.4% |
| Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled | F16 | 64.61 GiB | huihui-ai | 66.19 GiB | -2.4% |
| Qwen3.6-35B-A3B-NVFP4 | NVFP4 | 22.21 GiB | williamliao | 21.10 GiB | +5.3% |
| MOSS-Transcribe-Diarize | Q8_0 | 0.92 GiB | cstr | 1.31 GiB | -30.0% |
| MOSS-Transcribe-Diarize | F16 | 1.71 GiB | cstr | 1.70 GiB | +0.6% |
| Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled | F16 | 66.19 GiB | huihui-ai | 66.19 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
| Model | Quantizations | Smallest |
|---|---|---|
| Qwopus3.6-35B-A3B-v1 | 1 | 66.19 GiB |
| ced-base | 3 | 0.12 GiB |
| Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled | 2 | 64.61 GiB |
| Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 | 1 | 58.84 GiB |
| Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled | 1 | 66.19 GiB |
| parakeet-ctc-0.6b | 5 | 7.34 GiB |
| VibeVoice-ASR | 2 | 9.68 GiB |
| Qwen3.6-35B-A3B-NVFP4 | 1 | 22.21 GiB |
| DA3-BASE | 6 | 1.31 GiB |
| Carnice-Qwen3.6-MoE-35B-A3B | 1 | 66.19 GiB |
| LocateAnything-3B | 5 | 4.39 GiB |
| MOSS-Transcribe-Diarize | 7 | 0.50 GiB |
| Qwen3.5-35B-A3B-Uncensored-FernflowerAI-safetensors | 1 | 64.61 GiB |
| rfdetr | 12 | 0.03 GiB |
| MOSS-TTS-Local-Transformer-v1.5 | 3 | 5.56 GiB |
| Carnice-MoE-35B-A3B | 1 | 64.61 GiB |