README.md

Quantization made by Richard Erkhov.

[Github](https://github.com/RichardErkhov)

[Discord](https://discord.gg/pvy7H8DZMG)

[Request more models](https://github.com/RichardErkhov/quant_request)


LLaMA3-SFT - GGUF
- Model creator: https://huggingface.co/RLHFlow/
- Original model: https://huggingface.co/RLHFlow/LLaMA3-SFT/


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [LLaMA3-SFT.Q2_K.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q2_K.gguf) | Q2_K | 2.96GB |
| [LLaMA3-SFT.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.IQ3_XS.gguf) | IQ3_XS | 3.28GB |
| [LLaMA3-SFT.IQ3_S.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.IQ3_S.gguf) | IQ3_S | 3.43GB |
| [LLaMA3-SFT.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
| [LLaMA3-SFT.IQ3_M.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.IQ3_M.gguf) | IQ3_M | 3.52GB |
| [LLaMA3-SFT.Q3_K.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q3_K.gguf) | Q3_K | 3.74GB |
| [LLaMA3-SFT.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
| [LLaMA3-SFT.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
| [LLaMA3-SFT.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.IQ4_XS.gguf) | IQ4_XS | 4.18GB |
| [LLaMA3-SFT.Q4_0.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q4_0.gguf) | Q4_0 | 4.34GB |
| [LLaMA3-SFT.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.IQ4_NL.gguf) | IQ4_NL | 4.38GB |
| [LLaMA3-SFT.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q4_K_S.gguf) | Q4_K_S | 4.37GB |
| [LLaMA3-SFT.Q4_K.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q4_K.gguf) | Q4_K | 4.58GB |
| [LLaMA3-SFT.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q4_K_M.gguf) | Q4_K_M | 4.58GB |
| [LLaMA3-SFT.Q4_1.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q4_1.gguf) | Q4_1 | 4.78GB |
| [LLaMA3-SFT.Q5_0.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q5_0.gguf) | Q5_0 | 5.21GB |
| [LLaMA3-SFT.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
| [LLaMA3-SFT.Q5_K.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q5_K.gguf) | Q5_K | 5.34GB |
| [LLaMA3-SFT.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q5_K_M.gguf) | Q5_K_M | 5.34GB |
| [LLaMA3-SFT.Q5_1.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q5_1.gguf) | Q5_1 | 5.65GB |
| [LLaMA3-SFT.Q6_K.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q6_K.gguf) | Q6_K | 6.14GB |
| [LLaMA3-SFT.Q8_0.gguf](https://huggingface.co/RichardErkhov/RLHFlow_-_LLaMA3-SFT-gguf/blob/main/LLaMA3-SFT.Q8_0.gguf) | Q8_0 | 7.95GB |




Original model description:
---
library_name: transformers
tags: []
---

This is the SFT checkpoint used for the project [Online-RLHF](https://github.com/RLHFlow/Online-RLHF). Also check our [technical report here](https://arxiv.org/pdf/2405.07863). 

The model is trained from [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on a mixture of diverse open-source high-quality data for 1 epoch with detailed parameters in the report. It has not been trained by RLHF and can serve as a good starting point for the RLHF research.