Downloads · 30 days
128
10% of all-time downloads
RichardErkhov/cgato_-_L3-TheSpice-8b-v0.8.3-gguf
cgato_-_L3-TheSpice-8b-v0.8.3-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
128
10% of all-time downloads
All-time downloads
1.3K
Public
Repo size
92.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf108 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
L3-TheSpice-8b-v0.8.3 - GGUF
| Name | Quant method | Size |
|---|---|---|
| L3-TheSpice-8b-v0.8.3.Q2_K.gguf | Q2_K | 2.96GB |
| L3-TheSpice-8b-v0.8.3.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| L3-TheSpice-8b-v0.8.3.IQ3_S.gguf | IQ3_S | 3.43GB |
| L3-TheSpice-8b-v0.8.3.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| L3-TheSpice-8b-v0.8.3.IQ3_M.gguf | IQ3_M | 3.52GB |
| L3-TheSpice-8b-v0.8.3.Q3_K.gguf | Q3_K | 3.74GB |
| L3-TheSpice-8b-v0.8.3.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| L3-TheSpice-8b-v0.8.3.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| L3-TheSpice-8b-v0.8.3.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| L3-TheSpice-8b-v0.8.3.Q4_0.gguf | Q4_0 | 4.34GB |
| L3-TheSpice-8b-v0.8.3.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| L3-TheSpice-8b-v0.8.3.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| L3-TheSpice-8b-v0.8.3.Q4_K.gguf | Q4_K | 4.58GB |
| L3-TheSpice-8b-v0.8.3.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| L3-TheSpice-8b-v0.8.3.Q4_1.gguf | Q4_1 | 4.78GB |
| L3-TheSpice-8b-v0.8.3.Q5_0.gguf | Q5_0 | 5.21GB |
| L3-TheSpice-8b-v0.8.3.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| L3-TheSpice-8b-v0.8.3.Q5_K.gguf | Q5_K | 5.34GB |
| L3-TheSpice-8b-v0.8.3.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| L3-TheSpice-8b-v0.8.3.Q5_1.gguf | Q5_1 | 5.65GB |
| L3-TheSpice-8b-v0.8.3.Q6_K.gguf | Q6_K | 6.14GB |
| L3-TheSpice-8b-v0.8.3.Q8_0.gguf | Q8_0 | 7.95GB |
Now not overtrained and with the tokenizer fix to base llama3. Trained for 3 epochs.
The latest TheSpice, dipped in Mama Liz's LimaRP Oil. I've focused on making the model more flexible and provide a more unique experience. I'm still working on cleaning up my dataset, but I've shrunken it down a lot to focus on a "less is more" approach. This is ultimate a return to form of the way I used to train Thespis, with more of a focus on a small hand edited dataset.
Narration
If you request information on objects or characters in the scene, the model will narrate it to you. Most of the time, without moving the story forward.





If you're using Ooba in verbose mode as a server, you can check if you're console is logging something that looks like this.

{System Prompt}
Username: {Input}
BotName: {Response}
Username: {Input}
BotName: {Response}
All screenshots above were taken with the below SillyTavern Preset.
This is a roughly equivalent Kobold Horde Preset.
Please prompt responsibly and take anything outputted by any Language Model with a huge grain of salt. Thanks!