---
base_model:
- GreenerPastures/Bald-Beaver-8B
- allura-org/remnant-qwen3-8b
- allura-org/Q3-8B-Kintsugi
- Qwen/Qwen3-8B-Base
library_name: transformers
tags:
- mergekit
- merge
---

---

Potentially, this is one of the best 8B models for RP if you find the right settings that overcome the occasional repetitions.

(I'll say in advance — I'm not lucky with settings...)

---

# All-Q3-8B-RP-0625

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [DARE TIES](https://arxiv.org/abs/2311.03099) merge method using Qwen/Qwen3-8B-Base as a base.

### Models Merged

The following models were included in the merge:
* GreenerPastures/Bald-Beaver-8B
* allura-org/remnant-qwen3-8b
* allura-org/Q3-8B-Kintsugi

### Configuration

The following YAML configuration was used to produce this model:

```yaml
merge_method: dare_ties
base_model: Qwen/Qwen3-8B-Base
dtype: bfloat16
models:
  - model: GreenerPastures/Bald-Beaver-8B
    parameters:
      weight: 0.2
  - model: allura-org/Q3-8B-Kintsugi
    parameters:
      weight: 0.4
  - model: allura-org/remnant-qwen3-8b
    parameters:
      weight: 0.4
parameters:
  density: 0.35
```