Keiana-L3-Test5.6-8B-12

Keiana-L3-Test5.6-8B-12 is a merge of the following models using LazyMergekit:

Keep in mind that, this merged model isn't usually tested at the moment, which could benefit in vocabulary error.

• Kaoeiri/Keiana-L3-Test5.45-8B-10.5
• jeiku/Chaos_RP_l3_8B
• Sao10K/L3-Solana-8B-v1

🧩 Configuration

merge_method: model_stock
dtype: float16
base_model: Kaoeiri/Keiana-L3-Test5.4-8B-10
models:
  - model: Kaoeiri/Keiana-L3-Test5.45-8B-10.5
    parameters:
      weight: .126
      density: .216
  - model: jeiku/Chaos_RP_l3_8B
    parameters:
      weight: .128
      density: .256
  - model: Sao10K/L3-Solana-8B-v1
    parameters:
      weight: .16
      density: .12
parameters:
  normalize: true
  int8_mask: true

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Kaoeiri/Keiana-L3-Test5.6-8B-12"
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"])