Hugging Face's logo

R136a1
/
Ayam-2x8B-GGUF

GGUF
English
imatrix
Model card Files Community
Ayam-2x8B-GGUF
File size: 2,233 Bytes 
b9b98ca
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
 
---
license: cc-by-nc-4.0
language:
- en
---

`art by myself using AI`
![ayam](https://huggingface.co/R136a1/test1-2x8B/resolve/main/716495979429324710.png)

# Ayam 2x8B

> [!TIP]
> **GGUF** <br>
> Quantized using lastest llama.cpp as per writing. <br>
> Quantized using imatrix calculated from FP16 while using BF16 to create the quants to preserve accuracy. <br>

### Technical test:
| Quant     | PPL                  | VRAM    |
|-----------|----------------------|---------|
| FP16      | 3.7393 +/- 0.14377   | 22GB+   |
| Q8_0      | 3.7393 +/- 0.14381   | 14.8GB  |
| Q6_K      | 3.7283 +/- 0.14309   | 11.7GB  |
| Q5_K_M    | 3.7490 +/- 0.14440   | 10.4GB  |
| Q4_K_M    | 3.7263 +/- 0.14158   | 9.1GB   |
| Q4_K_S    | 3.7276 +/- 0.14139   | 8.5GB   |
| Q3_K_M    | 3.8198 +/- 0.14552   | 7.5GB   |

> [!NOTE]
> Perplexity test using llama.cpp/perplexity. <br>
> VRAM at full 8K context using Nvidia L4 GPU. <br>
> VRAM test using llama.cpp/server.

Another MoE, this time using L3.
<br>Recipe: Sao's [Stheno-v3.2](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.2) + L3 8B Instruct.

This model is intended for personal use but I think it's really good and worth sharing. Stheno-v3.2 is, as you probably know well, very good. In creative writing, RP and ERP it's far better than L3 Instruct and to be honest it's the best L3 finetunes I've tried so far so yeah I liked it very much. But while playing with it, I feel like the model is (a bit) dumber than L3 Instruct. It can't understand complex scenario well and confused in multi-char scenario, at least that what I was experiencing. So yeah, I tried to improve its intelligence while still preserving its creativity.

Why MoE not merge?
<br>Well... 2 model working together is always better than merging it into one. And (surprisingly) the result is far exceeded my expectations.
<br>And I think merging models sometimes can damage It's quality.

Testing condition (using [SillyTavern](https://github.com/SillyTavern/SillyTavern)):
<br>Context and Instruct: Llama 3 Instruct.
<br>Sampler:
```
Temperature : 1.15
MinP : 0.075
TopK : 50
Other is disabled.
```

## Switch
[BF16](https://huggingface.co/R136a1/Ayam-2x8B) - [GGUF](https://huggingface.co/R136a1/Ayam-2x8B-GGUF)