---
license: apache-2.0
language:
- en
pipeline_tag: text-generation
library_name: transformers
---

![GRaPE_Logo](https://cdn-uploads.huggingface.co/production/uploads/66960602f0ffd8e3a381106a/XjHkzctrE41e1qqJYeDzN.png)

_The **G**eneral **R**easoning **A**gent (for) **P**roject **E**xploration_

# The Model
| Attribute | Size | Modalities | Domain |
| :--- | :--- | :--- | :--- |
| **GRaPE Mini Writer** | 3B | Text + Image + Video in, Text out | Creative Writing Tasks |

***

# Capabilities

GRaPE Mini Writer was created for creative writing tasks, being more powerful than GRaPE Mini

> *Note: GRaPE Mini Writer doesn't think before responding*

***

# How to Run

I recommend using **LM Studio** for running GRaPE Models, and have generally found these sampling parameters to work best:

| Name | Value |
| :--- | :--- | 
| **Temperature** | 0.6 |
| **Top K Sampling** | 40 |
| **Repeat Penalty** | 1 |
| **Top P Sampling** | 0.85 |
| **Min P Sampling** | 0.05 |

***

# GRaPE Mini Writer as a Model

GRaPE Mini Writer was a model planned to be used as an expert in a "GRaPE Pro 1" model, which was eventually scrapped due to the aging architecture, riskiness of model upcycling, and the innsufficient compute that SLAI has to train it.

***

# Architecture

* GRaPE Mini Writer: Built on the GRaPE Mini's architecture

***

# Notes

The GRaPE Family started all the way back in August of 2025, meaning these models are severely out of date on architecture, and training data.

GRaPE 2 will come sooner than the GRaPE 1 family had, and will show multiple improvements.

There are no benchmarks for GRaPE 1 Models due to the costly nature of running them, as well as prioritization of newer models.

Updates for GRaPE 2 models will be posted here on Huggingface, as well as [Skinnertopia](https://www.skinnertopia.com/)