---
license: apache-2.0
language:
- en
tags:
- merge
- moe
---
![image/png](https://i.ibb.co/7k4j8Gm/icon10.png)
(Maybe i'll change the icon picture later.)

Experimental MoE, the idea is to have more active parameters than 7xX model would have and keep it's size lower than 20B.

This model has ~19.2B parameters.

[Exl2, 4.0 bpw](https://huggingface.co/xxx777xxxASD/PrimaMonarch-EroSumika-2x10.7B-128k-bpw-4.0) (Fits in 12GB VRAM/16k context/4-bit cache)

[Exl2, 6.0 bpw](https://huggingface.co/xxx777xxxASD/PrimaMonarch-EroSumika-2x10.7B-128k-bpw-6.0)

[GGUF](https://huggingface.co/xxx777xxxASD/PrimaMonarch-EroSumika-2x10.7B-128k-GGUF)

### Base model (self merge)
```
slices:
  - sources:
    - model: MistralInstruct-v0.2-128k
      layer_range: [0, 24]
  - sources:
    - model: MistralInstruct-v0.2-128k
      layer_range: [8, 24]
  - sources:
    - model: MistralInstruct-v0.2-128k
      layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16
```

### First expert ("sandwich" merge)
[xxx777xxxASD/PrimaSumika-10.7B-128k](https://huggingface.co/xxx777xxxASD/PrimaSumika-10.7B-128k)
```
slices:
  - sources:
    - model: EroSumika-128k
      layer_range: [0, 24]
  - sources:
    - model: Prima-Lelantacles-128k
      layer_range: [8, 24]
  - sources:
    - model: EroSumika-128k
      layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16

```

### Second expert ("sandwich" merge)
```
slices:
  - sources:
    - model: AlphaMonarch-7B-128k
      layer_range: [0, 24]
  - sources:
    - model: NeuralHuman-128k
      layer_range: [8, 24]
  - sources:
    - model: AlphaMonarch-7B-128k
      layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16
```

Each 128k model is a slerp merge with [Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context](https://huggingface.co/Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context)

## Models used

- [localfultonextractor/Erosumika-7B](https://huggingface.co/localfultonextractor/Erosumika-7B)
- [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
- [Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context](https://huggingface.co/Epiculous/Fett-uccine-Long-Noodle-7B-120k-Context)
- [mlabonne/AlphaMonarch-7B](https://huggingface.co/mlabonne/AlphaMonarch-7B)
- [Nitral-AI/Prima-LelantaclesV6-7b](https://huggingface.co/Nitral-AI/Prima-LelantaclesV6-7b)
- [NeuralNovel/Mistral-7B-Instruct-v0.2-Neural-Story](https://huggingface.co/NeuralNovel/Mistral-7B-Instruct-v0.2-Neural-Story)
- [valine/MoreHuman](https://huggingface.co/valine/MoreHuman)