File size: 1,552 Bytes
954964c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
---
base_model: Goekdeniz-Guelmez/Josiefied-Qwen3-8B-abliterated-v1
---

## [EXL3](https://github.com/turboderp-org/exllamav3) quantization of [Josiefied-Qwen3-8B-abliterated-v1](https://huggingface.co/Goekdeniz-Guelmez/Josiefied-Qwen3-8B-abliterated-v1)

### HumanEval (argmax)

| Model                                                                                                                          | Q4   | Q6   | Q8   | FP16 |
| ------------------------------------------------------------------------------------------------------------------------------ | ---- | ---- | ---- | ---- |
| [Josiefied-Qwen3-8B-abliterated-v1-exl3-4bpw](https://huggingface.co/isogen/Josiefied-Qwen3-8B-abliterated-v1-exl3-4bpw)       | 31.7 | 37.8 | 36.6 | 35.4 |
| [Josiefied-Qwen3-8B-abliterated-v1-exl3-6bpw](https://huggingface.co/isogen/Josiefied-Qwen3-8B-abliterated-v1-exl3-6bpw)       | 32.9 | 36.6 | 37.2 | 37.8 |
| [Josiefied-Qwen3-8B-abliterated-v1-exl3-8bpw-h8](https://huggingface.co/isogen/Josiefied-Qwen3-8B-abliterated-v1-exl3-8bpw-h8) | 34.1 | 36.0 | 34.1 | 35.4 |
| [Qwen3-8B-exl3 4bpw](https://huggingface.co/turboderp/Qwen3-8B-exl3)                                                           | 44.5 | 45.7 | 47.6 | 48.2 |
| [Qwen3-8B-exl3 6bpw](https://huggingface.co/turboderp/Qwen3-8B-exl3)                                                           | 37.8 | 43.3 | 46.3 | 45.1 |
| [Qwen3-8B-exl3 8bpw-h8](https://huggingface.co/turboderp/Qwen3-8B-exl3)                                                        | 29.9 | 37.8 | 37.8 | 39.0 |