Big-Firefly

Big-Firefly is a merge of the following models using LazyMergekit:

• TheDrummer/Cydonia-24B-V4.3
• Doctor-Shotgun/MS3.2-24B-Magnum-Diamond
• zerofata/MS3.2-PaintedFantasy-v3-24B

🧩 Configuration

models:
  - model: TheDrummer/Cydonia-24B-V4.3
    parameters:
      weight: 0.45
  - model: Doctor-Shotgun/MS3.2-24B-Magnum-Diamond
    parameters:
      weight: 0.30
  - model: zerofata/MS3.2-PaintedFantasy-v3-24B
    parameters:
      weight: 0.25
merge_method: task_arithmetic
base_model: TheDrummer/Cydonia-24B-v4.3
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Guilherme34/Big-Firefly"
messages = [{"role": "user", "content": "What is a large language model?"}]

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"])