Quantization made by Richard Erkhov.

Github

Discord

Request more models

gpt2o-chatbot-02 - GGUF

• Model creator: https://huggingface.co/styalai/
• Original model: https://huggingface.co/styalai/gpt2o-chatbot-02/

Name	Quant method	Size
gpt2o-chatbot-02.Q2_K.gguf	Q2_K	0.84GB
gpt2o-chatbot-02.IQ3_XS.gguf	IQ3_XS	0.84GB
gpt2o-chatbot-02.IQ3_S.gguf	IQ3_S	0.84GB
gpt2o-chatbot-02.Q3_K_S.gguf	Q3_K_S	0.84GB
gpt2o-chatbot-02.IQ3_M.gguf	IQ3_M	0.91GB
gpt2o-chatbot-02.Q3_K.gguf	Q3_K	0.97GB
gpt2o-chatbot-02.Q3_K_M.gguf	Q3_K_M	0.97GB
gpt2o-chatbot-02.Q3_K_L.gguf	Q3_K_L	1.03GB
gpt2o-chatbot-02.IQ4_XS.gguf	IQ4_XS	0.9GB
gpt2o-chatbot-02.Q4_0.gguf	Q4_0	0.91GB
gpt2o-chatbot-02.IQ4_NL.gguf	IQ4_NL	0.91GB
gpt2o-chatbot-02.Q4_K_S.gguf	Q4_K_S	1.04GB
gpt2o-chatbot-02.Q4_K.gguf	Q4_K	1.11GB
gpt2o-chatbot-02.Q4_K_M.gguf	Q4_K_M	1.11GB
gpt2o-chatbot-02.Q4_1.gguf	Q4_1	1.0GB
gpt2o-chatbot-02.Q5_0.gguf	Q5_0	1.09GB
gpt2o-chatbot-02.Q5_K_S.gguf	Q5_K_S	1.15GB
gpt2o-chatbot-02.Q5_K.gguf	Q5_K	1.29GB
gpt2o-chatbot-02.Q5_K_M.gguf	Q5_K_M	1.29GB
gpt2o-chatbot-02.Q5_1.gguf	Q5_1	1.18GB
gpt2o-chatbot-02.Q6_K.gguf	Q6_K	1.52GB
gpt2o-chatbot-02.Q8_0.gguf	Q8_0	1.63GB

Original model description:

tags:

• autotrain
• text-generation-inference
• text-generation
• peft library_name: transformers base_model: openai-community/gpt2-xl 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)