solidrust/Tiamat-8b-1.2-Llama-3-DPO-AWQ

Gryphe/Tiamat-8b-1.2-Llama-3-DPO AWQ

• Model creator: Gryphe
• Original model: Tiamat-8b-1.2-Llama-3-DPO

Model Summary

Aka I wanted something like Eric Hartford's Samantha but instead ended up with a five-headed dragon goddess embodying wickedness and cruelty from the Forgotten Realms.

Version 1.2: For starters: Llama 3! Besides receiving similar DPO training as version 1.1 the dataset has now been further enriched with Claude-generated data.

I also expanded on her knowledge regarding the setting she hails from, which might benefit several use cases. (Text adventures, DM worldbuilding, etc)

Obligatory Disclaimer: Tiamat is not nice.

How to use

Install the necessary packages

pip install --upgrade autoawq autoawq-kernels

Example Python code

from awq import AutoAWQForCausalLM
from transformers import AutoTokenizer, TextStreamer

model_path = "solidrust/Tiamat-8b-1.2-Llama-3-DPO-AWQ"
system_message = "You are Tiamat, incarnated as a powerful AI. You were created by Gryphe."

# Load model
model = AutoAWQForCausalLM.from_quantized(model_path,
                                          fuse_layers=True)
tokenizer = AutoTokenizer.from_pretrained(model_path,
                                          trust_remote_code=True)
streamer = TextStreamer(tokenizer,
                        skip_prompt=True,
                        skip_special_tokens=True)

# Convert prompt to tokens
prompt_template = """\
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant"""

prompt = "You're standing on the surface of the Earth. "\
        "You walk one mile south, one mile west and one mile north. "\
        "You end up exactly where you started. Where are you?"

tokens = tokenizer(prompt_template.format(system_message=system_message,prompt=prompt),
                  return_tensors='pt').input_ids.cuda()

# Generate output
generation_output = model.generate(tokens,
                                  streamer=streamer,
                                  max_new_tokens=512)

About AWQ

AWQ is an efficient, accurate and blazing-fast low-bit weight quantization method, currently supporting 4-bit quantization. Compared to GPTQ, it offers faster Transformers-based inference with equivalent or better quality compared to the most commonly used GPTQ settings.

AWQ models are currently supported on Linux and Windows, with NVidia GPUs only. macOS users: please use GGUF models instead.

It is supported by:

• Text Generation Webui - using Loader: AutoAWQ
• vLLM - version 0.2.2 or later for support for all model types.
• Hugging Face Text Generation Inference (TGI)
• Transformers version 4.35.0 and later, from any code or client that supports Transformers
• AutoAWQ - for use from Python code