---
license: other
license_link: LICENSE
base_model:
- Nitral-AI/CaptainErisNebula-12B-Chimera-v1.1
tags:
- text-generation-inference
- testing
- mistral
- chatml
---

> [!TIP]
> # **GGUF quants for [**Nitral-AI/CaptainErisNebula-12B-Chimera-v1.1**](https://huggingface.co/Nitral-AI/CaptainErisNebula-12B-Chimera-v1.1)'s recipe.**

> [!IMPORTANT]
> **Author recommended initial SillyTavern presets:** <br>
> - [Chimera: v1.1](https://huggingface.co/Nitral-AI/CaptainErisNebula-12B-Chimera-v1.1/tree/main/ST)
>   - [(backup)(633788596ef1810cc9ea9abd375bbda3e568bd37)](https://huggingface.co/Lewdiculous/CaptainErisNebula-12B-Chimera-v1.1-GGUF-IQ-Imatrix/tree/main/sillytavern)

![https://iili.io/KKtCMf2.md.jpg](https://iili.io/KKtCMf2.md.jpg)

<hr>

> [!NOTE]
> ## This is an improvement on the previous experimental version. <br>
> - Not "chaotic", and at a usable size for most people seeking to perform inference locally with good speeds. <br>
> - The model does not show excessive alignment, so it should be good for most scenarios/writing situations. <br> <br>
> - Feel free to use some light system prompting to nudge it out of a blocker if needed. <br> <br>
> - It does well in adhering to characters and instructions. <br> <br>

<hr>

Thank you so much, "crazy chef" and "mad scientist", **Nitral**!

<!-- ![red nebula inspired model card image](https://iili.io/KJuO2Ie.png) -->

<br>

```text
# Using the latest llama.cpp ...
release version at the time: b6258.
# Imatrix was based on the full ...
FP16 precision GGUF.

START: BF16 HuggingFace Model
↓
(1) Conversion to Full-Precision GGUF
↓
FP16 GGUF (for Calibration Imatrix)
BF16 GGUF (for Quantization)
↓
(2) Generate Imatrix (from FP16 GGUF)
↓
imatrix.fp16.gguf
↓
(3) Quantize with Imatrix (using BF16 GGUF)
↓
Final Quantized GGUF Models
↓
END
```