84 lines
4.3 KiB
Markdown
84 lines
4.3 KiB
Markdown
---
|
|
license: cc-by-sa-4.0
|
|
language:
|
|
- en
|
|
base_model:
|
|
- Qwen/Qwen3-1.7B
|
|
pipeline_tag: text-generation
|
|
---
|
|
|
|
# GGUF Files for FoxyzGPT-X1.1-1.7B
|
|
|
|
These are the GGUF files for [Flexan/FoxyzGPT-X1.1-1.7B](https://huggingface.co/Flexan/FoxyzGPT-X1.1-1.7B).
|
|
|
|
## Downloads
|
|
|
|
| GGUF Link | Quantization | Description |
|
|
| ---- | ----- | ----------- |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q2_K.gguf) | Q2_K | Lowest quality |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q3_K_S.gguf) | Q3_K_S | |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.IQ3_S.gguf) | IQ3_S | Integer quant, preferable over Q3_K_S |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.IQ3_M.gguf) | IQ3_M | Integer quant |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q3_K_M.gguf) | Q3_K_M | |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q3_K_L.gguf) | Q3_K_L | |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.IQ4_XS.gguf) | IQ4_XS | Integer quant |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q4_K_S.gguf) | Q4_K_S | Fast with good performance |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q4_K_M.gguf) | Q4_K_M | **Recommended:** Perfect mix of speed and performance |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q5_K_S.gguf) | Q5_K_S | |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q5_K_M.gguf) | Q5_K_M | |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q6_K.gguf) | Q6_K | Very good quality |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.Q8_0.gguf) | Q8_0 | Best quality |
|
|
| [Download](https://huggingface.co/Flexan/Flexan-FoxyzGPT-X1.1-1.7B-GGUF/resolve/main/FoxyzGPT-X1.1-1.7B.f16.gguf) | f16 | Full precision, don't bother; use a quant |
|
|
|
|
# FoxyzGPT X1.1 1.7B
|
|
|
|
## Description
|
|
|
|
FoxyzGPT X1.1 1.7B is an instruct LLM consisting of 1.7B parameters trained to talk in a human conversational manner. It does not support reasoning nor tool-calling (although the base model does).
|
|
The model was LoRA fine-tuned with [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) as base model.
|
|
|
|
This model is trained on a private dataset provided by [Foxyz](https://huggingface.co/Foxyz). The model has adapted this persona and will not be accurate at answering questions at all.
|
|
|
|
## Chat Format
|
|
|
|
FoxyzGPT X1.1 1.7B uses the ChatML format, e.g.:
|
|
```text
|
|
<|im_start|>system
|
|
System message<|im_end|>
|
|
<|im_start|>user
|
|
User prompt<|im_end|>
|
|
<|im_start|>assistant
|
|
Assistant response<|im_end|>
|
|
```
|
|
|
|
## Usage
|
|
|
|
This model is trained on one system prompt only. Therefore, it is recommended to use the one stated here:
|
|
```text
|
|
You are Foxyz (username: foxyz9248) and you are talking to `<username>` on Discord in a direct message. Use `~>` to signal the start of a new message (you can send multiple messages this way). Use silly language.
|
|
```
|
|
|
|
You should replace `<username>` with your username. You can obviously use this LLM for platforms other than Discord, but in the system prompt it's best to explicitly state the user is on Discord (the model has not been trained with other system prompt variations besides the username).
|
|
|
|
The assistant response has the following format:
|
|
```text
|
|
<|im_start|>assistant
|
|
~> HLELOO!!!
|
|
~> howa reyou :D<|im_end|>
|
|
```
|
|
|
|
Note that the prompt should be formatted differently. The prompt is composed as a list of messages using the `~>` arrow notation. `~>` marks the start of a new message. This approach allows you to send multiple messages while still allowing multi-line without using multiple `user` roles:
|
|
```text
|
|
<|im_start|>user
|
|
~> my first message
|
|
~> my second message
|
|
~> this is
|
|
a message
|
|
that has 4
|
|
lines
|
|
~> and this is my fourth message<|im_end|>
|
|
```
|
|
|
|
## Datasets
|
|
|
|
1. **Private dataset** *3.1k chats* |