Quantization made by Richard Erkhov.

Wernicke-7B-v9 - GGUF

Model creator: https://huggingface.co/CultriX/
Original model: https://huggingface.co/CultriX/Wernicke-7B-v9/

Name	Quant method	Size
Wernicke-7B-v9.Q2_K.gguf	Q2_K	2.53GB
Wernicke-7B-v9.IQ3_XS.gguf	IQ3_XS	2.81GB
Wernicke-7B-v9.IQ3_S.gguf	IQ3_S	2.96GB
Wernicke-7B-v9.Q3_K_S.gguf	Q3_K_S	2.95GB
Wernicke-7B-v9.IQ3_M.gguf	IQ3_M	3.06GB
Wernicke-7B-v9.Q3_K.gguf	Q3_K	3.28GB
Wernicke-7B-v9.Q3_K_M.gguf	Q3_K_M	3.28GB
Wernicke-7B-v9.Q3_K_L.gguf	Q3_K_L	3.56GB
Wernicke-7B-v9.IQ4_XS.gguf	IQ4_XS	3.67GB
Wernicke-7B-v9.Q4_0.gguf	Q4_0	3.83GB
Wernicke-7B-v9.IQ4_NL.gguf	IQ4_NL	3.87GB
Wernicke-7B-v9.Q4_K_S.gguf	Q4_K_S	3.86GB
Wernicke-7B-v9.Q4_K.gguf	Q4_K	4.07GB
Wernicke-7B-v9.Q4_K_M.gguf	Q4_K_M	4.07GB
Wernicke-7B-v9.Q4_1.gguf	Q4_1	4.24GB
Wernicke-7B-v9.Q5_0.gguf	Q5_0	4.65GB
Wernicke-7B-v9.Q5_K_S.gguf	Q5_K_S	4.65GB
Wernicke-7B-v9.Q5_K.gguf	Q5_K	4.78GB
Wernicke-7B-v9.Q5_K_M.gguf	Q5_K_M	4.78GB
Wernicke-7B-v9.Q5_1.gguf	Q5_1	5.07GB
Wernicke-7B-v9.Q6_K.gguf	Q6_K	5.53GB
Wernicke-7B-v9.Q8_0.gguf	Q8_0	7.17GB

Original model description:

tags: - merge - mergekit - lazymergekit - FelixChao/WestSeverus-7B-DPO-v2 - CultriX/Wernicke-7B-v8 - vanillaOVO/supermario_v2 base_model: - FelixChao/WestSeverus-7B-DPO-v2 - CultriX/Wernicke-7B-v8 - vanillaOVO/supermario_v2 license: apache-2.0

Edit:

Best Wernicke Model yet.
Benchmark Results: https://huggingface.co/spaces/CultriX/Yet_Another_LLM_Leaderboard

Wernicke-7B-v9

Wernicke-7B-v9 is a merge of the following models using LazyMergekit:

🧩 Configuration

models:
  - model: FelixChao/WestSeverus-7B-DPO-v2
    parameters:
      density: 0.50
      weight: 0.35
  - model: CultriX/Wernicke-7B-v8
    parameters:
      density: 0.50
      weight: 0.35
  - model: vanillaOVO/supermario_v2
    parameters:
      density: 0.50
      weight: 0.30
merge_method: dare_ties
base_model: FelixChao/WestSeverus-7B-DPO-v2
parameters:
  int8_mask: true
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "CultriX/Wernicke-7B-v9"
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"])