RichardErkhov/numfa_-_numfa_v2-3b-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

numfa_v2-3b - GGUF

• Model creator: https://huggingface.co/numfa/
• Original model: https://huggingface.co/numfa/numfa_v2-3b/

Name	Quant method	Size
numfa_v2-3b.Q2_K.gguf	Q2_K	2.0GB
numfa_v2-3b.IQ3_XS.gguf	IQ3_XS	2.0GB
numfa_v2-3b.IQ3_S.gguf	IQ3_S	2.0GB
numfa_v2-3b.Q3_K_S.gguf	Q3_K_S	2.0GB
numfa_v2-3b.IQ3_M.gguf	IQ3_M	2.07GB
numfa_v2-3b.Q3_K.gguf	Q3_K	2.15GB
numfa_v2-3b.Q3_K_M.gguf	Q3_K_M	2.15GB
numfa_v2-3b.Q3_K_L.gguf	Q3_K_L	2.22GB
numfa_v2-3b.IQ4_XS.gguf	IQ4_XS	2.02GB
numfa_v2-3b.Q4_0.gguf	Q4_0	2.0GB
numfa_v2-3b.IQ4_NL.gguf	IQ4_NL	2.02GB
numfa_v2-3b.Q4_K_S.gguf	Q4_K_S	2.41GB
numfa_v2-3b.Q4_K.gguf	Q4_K	2.57GB
numfa_v2-3b.Q4_K_M.gguf	Q4_K_M	2.57GB
numfa_v2-3b.Q4_1.gguf	Q4_1	2.2GB
numfa_v2-3b.Q5_0.gguf	Q5_0	2.4GB
numfa_v2-3b.Q5_K_S.gguf	Q5_K_S	2.6GB
numfa_v2-3b.Q5_K.gguf	Q5_K	2.74GB
numfa_v2-3b.Q5_K_M.gguf	Q5_K_M	2.74GB
numfa_v2-3b.Q5_1.gguf	Q5_1	2.6GB
numfa_v2-3b.Q6_K.gguf	Q6_K	3.6GB
numfa_v2-3b.Q8_0.gguf	Q8_0	3.6GB

Original model description:

license: apache-2.0 language: - en library_name: transformers

NumFa v2 (3B)

NumFa v2 3B is a LLM pretrained that has 1B.

Base model: TinyLLama

For testing only

Model Details

Model Description

The model was trained by TPU.

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

• Developed by: NumFa
• Model type: text-generation
• Language(s) (NLP): English
• License: apache-2.0

Out-of-Scope Use

Math, Coding, and other language

Bias, Risks, and Limitations

The model can has a bias from dataset. Use at your own risks!

How to Get Started with the Model

Use the code below to get started with the model.

Example

# !pip install accelerate sentencepiece transformers bitsandbytes
import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="numfa/numfa_v2-3b", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating

outputs = pipe("test is", max_new_tokens=300, do_sample=True, temperature=0.9, top_k=50, top_p=0.95, no_repeat_ngram_size=2,typical_p=1.)
print(outputs[0]["generated_text"])