athirdpath_-_Hestia-20b-gguf

RichardErkhov/athirdpath_-_Hestia-20b-gguf

athirdpath_-_Hestia-20b-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

4% of all-time downloads

All-time downloads

302

Public

Repo size

226 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.gguf262 GB · 100%

At a glance

Access: Public
Created: Jun 2, 2024
Updated: Jun 2, 2024
SHA: 727e2ab6

Model card

From the Hugging Face model README

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Hestia-20b - GGUF

Model creator: https://huggingface.co/athirdpath/
Original model: https://huggingface.co/athirdpath/Hestia-20b/

Name	Quant method	Size
Hestia-20b.Q2_K.gguf	Q2_K	6.91GB
Hestia-20b.IQ3_XS.gguf	IQ3_XS	7.63GB
Hestia-20b.IQ3_S.gguf	IQ3_S	8.06GB
Hestia-20b.Q3_K_S.gguf	Q3_K_S	8.06GB
Hestia-20b.IQ3_M.gguf	IQ3_M	8.53GB
Hestia-20b.Q3_K.gguf	Q3_K	9.04GB
Hestia-20b.Q3_K_M.gguf	Q3_K_M	9.04GB
Hestia-20b.Q3_K_L.gguf	Q3_K_L	9.9GB
Hestia-20b.IQ4_XS.gguf	IQ4_XS	10.01GB
Hestia-20b.Q4_0.gguf	Q4_0	10.52GB
Hestia-20b.IQ4_NL.gguf	IQ4_NL	10.57GB
Hestia-20b.Q4_K_S.gguf	Q4_K_S	10.59GB
Hestia-20b.Q4_K.gguf	Q4_K	11.22GB
Hestia-20b.Q4_K_M.gguf	Q4_K_M	11.22GB
Hestia-20b.Q4_1.gguf	Q4_1	11.67GB
Hestia-20b.Q5_0.gguf	Q5_0	12.83GB
Hestia-20b.Q5_K_S.gguf	Q5_K_S	12.83GB
Hestia-20b.Q5_K.gguf	Q5_K	13.18GB
Hestia-20b.Q5_K_M.gguf	Q5_K_M	13.18GB
Hestia-20b.Q5_1.gguf	Q5_1	13.98GB
Hestia-20b.Q6_K.gguf	Q6_K	15.28GB
Hestia-20b.Q8_0.gguf	Q8_0	19.79GB

Original model description:

license: cc-by-nc-4.0

This is a task_arithmetic merge of Harmonia (my 20b faux base model) with Noromaid and my LORA-glued Nethena. Solidly outperforms Harmonia.

merge_method: task_arithmetic

base_model: athirdpath/Harmonia-20b

models:

model: athirdpath/Harmonia-20b
model: NeverSleep/Noromaid-20b-v0.1.1
- parameters: weight: 0.25
model: athirdpath/Nethena-20b-Glued
- parameters: weight: 0.2

dtype: float16

Thanks to Undi95 for pioneering the 20B recipe, and for most of the models involved.

Created: Jun 2, 2024
Updated: Jun 2, 2024