---
base_model: Goekdeniz-Guelmez/Josiefied-Qwen3-1.7B-abliterated-v1
---

[EXL3](https://github.com/turboderp-org/exllamav3) quantization of [Josiefied-Qwen3-1.7B-abliterated-v1](https://huggingface.co/Goekdeniz-Guelmez/Josiefied-Qwen3-1.7B-abliterated-v1), 8 bits per weight, including output layers.

### HumanEval (argmax)

| Model                                                                                                                              | Q4   | Q6    | Q8    | FP16  |
| ---------------------------------------------------------------------------------------------------------------------------------- | ---- | ----- | ----- | ----- |
| [Josiefied-Qwen3-1.7B-abliterated-v1-exl3-8bpw-h8](https://huggingface.co/isogen/Josiefied-Qwen3-1.7B-abliterated-v1-exl3-8bpw-h8) | 0.0% | 62.2% | 61.6% | 62.8% |
| [Qwen3-1.7B-exl3-8bpw-h8](https://huggingface.co/isogen/Qwen3-1.7B-exl3-8bpw-h8)                                                   | 0.0% | 70.7% | 68.3% | 68.9% |
| [Qwen3-1.7B-Base-exl3-8bpw-h8](https://huggingface.co/isogen/Qwen3-1.7B-Base-exl3-8bpw-h8)                                         | 0.0% | 66.5% | 70.7% | 70.1% |