Quantization made by Richard Erkhov.

CodeMaster-v1-9b - GGUF

Model creator: https://huggingface.co/KingNish/
Original model: https://huggingface.co/KingNish/CodeMaster-v1-9b/

Name	Quant method	Size
CodeMaster-v1-9b.Q2_K.gguf	Q2_K	3.19GB
CodeMaster-v1-9b.IQ3_XS.gguf	IQ3_XS	3.52GB
CodeMaster-v1-9b.IQ3_S.gguf	IQ3_S	3.72GB
CodeMaster-v1-9b.Q3_K_S.gguf	Q3_K_S	3.72GB
CodeMaster-v1-9b.IQ3_M.gguf	IQ3_M	3.93GB
CodeMaster-v1-9b.Q3_K.gguf	Q3_K	4.16GB
CodeMaster-v1-9b.Q3_K_M.gguf	Q3_K_M	4.16GB
CodeMaster-v1-9b.Q3_K_L.gguf	Q3_K_L	4.55GB
CodeMaster-v1-9b.IQ4_XS.gguf	IQ4_XS	4.61GB
CodeMaster-v1-9b.Q4_0.gguf	Q4_0	4.84GB
CodeMaster-v1-9b.IQ4_NL.gguf	IQ4_NL	4.86GB
CodeMaster-v1-9b.Q4_K_S.gguf	Q4_K_S	4.87GB
CodeMaster-v1-9b.Q4_K.gguf	Q4_K	5.16GB
CodeMaster-v1-9b.Q4_K_M.gguf	Q4_K_M	5.16GB
CodeMaster-v1-9b.Q4_1.gguf	Q4_1	5.36GB
CodeMaster-v1-9b.Q5_0.gguf	Q5_0	5.89GB
CodeMaster-v1-9b.Q5_K_S.gguf	Q5_K_S	5.89GB
CodeMaster-v1-9b.Q5_K.gguf	Q5_K	6.06GB
CodeMaster-v1-9b.Q5_K_M.gguf	Q5_K_M	6.06GB
CodeMaster-v1-9b.Q5_1.gguf	Q5_1	6.41GB
CodeMaster-v1-9b.Q6_K.gguf	Q6_K	7.01GB
CodeMaster-v1-9b.Q8_0.gguf	Q8_0	9.07GB

Original model description:

tags: - merge - mergekit - lazymergekit - KingNish/CodeMaster-v1-7b base_model: - KingNish/CodeMaster-v1-7b - KingNish/CodeMaster-v1-7b license: mit pipeline_tag: text-generation

CodeMaster-v1-9b

CodeMaster-v1-9b is a merge of the following models using LazyMergekit:

🧩 Configuration

slices:
  - sources:
    - model: KingNish/CodeMaster-v1-7b
      layer_range: [0, 22]
  - sources:
    - model: KingNish/CodeMaster-v1-7b
      layer_range: [10, 32]
merge_method: passthrough
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "KingNish/CodeMaster-v1-9b"
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=8192, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])