Downloads · 30 days
276
14% of all-time downloads
RichardErkhov/itsnebulalol_-_Llama-3.2-Nemotron-3B-gguf
itsnebulalol_-_Llama-3.2-Nemotron-3B-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
276
14% of all-time downloads
All-time downloads
1.9K
Public
Repo size
33.5 GB
Likes
0
Public
Click a slice to open those files.
.gguf39.5 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
Llama-3.2-Nemotron-3B - GGUF
| Name | Quant method | Size |
|---|---|---|
| Llama-3.2-Nemotron-3B.Q2_K.gguf | Q2_K | 1.27GB |
| Llama-3.2-Nemotron-3B.Q3_K_S.gguf | Q3_K_S | 1.44GB |
| Llama-3.2-Nemotron-3B.Q3_K.gguf | Q3_K | 1.57GB |
| Llama-3.2-Nemotron-3B.Q3_K_M.gguf | Q3_K_M | 1.57GB |
| Llama-3.2-Nemotron-3B.Q3_K_L.gguf | Q3_K_L | 1.69GB |
| Llama-3.2-Nemotron-3B.IQ4_XS.gguf | IQ4_XS | 1.71GB |
| Llama-3.2-Nemotron-3B.Q4_0.gguf | Q4_0 | 1.79GB |
| Llama-3.2-Nemotron-3B.IQ4_NL.gguf | IQ4_NL | 1.79GB |
| Llama-3.2-Nemotron-3B.Q4_K_S.gguf | Q4_K_S | 1.8GB |
| Llama-3.2-Nemotron-3B.Q4_K.gguf | Q4_K | 1.88GB |
| Llama-3.2-Nemotron-3B.Q4_K_M.gguf | Q4_K_M | 1.88GB |
| Llama-3.2-Nemotron-3B.Q4_1.gguf | Q4_1 | 1.95GB |
| Llama-3.2-Nemotron-3B.Q5_0.gguf | Q5_0 | 2.11GB |
| Llama-3.2-Nemotron-3B.Q5_K_S.gguf | Q5_K_S | 2.11GB |
| Llama-3.2-Nemotron-3B.Q5_K.gguf | Q5_K | 2.16GB |
| Llama-3.2-Nemotron-3B.Q5_K_M.gguf | Q5_K_M | 2.16GB |
| Llama-3.2-Nemotron-3B.Q5_1.gguf | Q5_1 | 2.28GB |
| Llama-3.2-Nemotron-3B.Q6_K.gguf | Q6_K | 2.46GB |
| Llama-3.2-Nemotron-3B.Q8_0.gguf | Q8_0 | 3.19GB |
license: llama3.2 language:
This is a finetune of meta-llama/Llama-3.2-3B (specifically, unsloth/Llama-3.2-3B-bnb-4bit).
It was trained on the nvidia/HelpSteer2 dataset, similar to nvidia/Llama-3.1-Nemotron-70B-Instruct-HF, using Unsloth.
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "itsnebulalol/Llama-3.2-Nemotron-3B"
messages = [{"role": "user", "content": "How many r in strawberry?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.