Quantization made by Richard Erkhov.

r-3b-tesla - GGUF

Model creator: https://huggingface.co./winstcha/
Original model: https://huggingface.co./winstcha/r-3b-tesla/

Name	Quant method	Size
r-3b-tesla.Q2_K.gguf	Q2_K	1.19GB
r-3b-tesla.IQ3_XS.gguf	IQ3_XS	1.3GB
r-3b-tesla.IQ3_S.gguf	IQ3_S	1.36GB
r-3b-tesla.Q3_K_S.gguf	Q3_K_S	1.35GB
r-3b-tesla.IQ3_M.gguf	IQ3_M	1.39GB
r-3b-tesla.Q3_K.gguf	Q3_K	1.48GB
r-3b-tesla.Q3_K_M.gguf	Q3_K_M	1.48GB
r-3b-tesla.Q3_K_L.gguf	Q3_K_L	1.59GB
r-3b-tesla.IQ4_XS.gguf	IQ4_XS	1.63GB
r-3b-tesla.Q4_0.gguf	Q4_0	1.7GB
r-3b-tesla.IQ4_NL.gguf	IQ4_NL	1.71GB
r-3b-tesla.Q4_K_S.gguf	Q4_K_S	1.71GB
r-3b-tesla.Q4_K.gguf	Q4_K	1.8GB
r-3b-tesla.Q4_K_M.gguf	Q4_K_M	1.8GB
r-3b-tesla.Q4_1.gguf	Q4_1	1.86GB
r-3b-tesla.Q5_0.gguf	Q5_0	2.02GB
r-3b-tesla.Q5_K_S.gguf	Q5_K_S	2.02GB
r-3b-tesla.Q5_K.gguf	Q5_K	2.07GB
r-3b-tesla.Q5_K_M.gguf	Q5_K_M	2.07GB
r-3b-tesla.Q5_1.gguf	Q5_1	2.18GB
r-3b-tesla.Q6_K.gguf	Q6_K	2.36GB
r-3b-tesla.Q8_0.gguf	Q8_0	3.06GB

Original model description:

tags: - autotrain - text-generation-inference - text-generation - peft library_name: transformers base_model: Qwen/Qwen2.5-3B-Instruct widget: - messages: - role: user content: What is your favorite condiment? license: other

Model Trained Using AutoTrain

This model was trained using AutoTrain. For more information, please visit AutoTrain.

Usage


from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = "PATH_TO_THIS_REPO"

tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(
    model_path,
    device_map="auto",
    torch_dtype='auto'
).eval()

# Prompt content: "hi"
messages = [
    {"role": "user", "content": "hi"}
]

input_ids = tokenizer.apply_chat_template(conversation=messages, tokenize=True, add_generation_prompt=True, return_tensors='pt')
output_ids = model.generate(input_ids.to('cuda'))
response = tokenizer.decode(output_ids[0][input_ids.shape[1]:], skip_special_tokens=True)

# Model response: "Hello! How can I assist you today?"
print(response)