Quantization made by Richard Erkhov.

NinjaDolphin-7B - GGUF

Model creator: https://huggingface.co/FelixChao/
Original model: https://huggingface.co/FelixChao/NinjaDolphin-7B/

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

Original model description:

license: apache-2.0 tags: - merge - beowolx/CodeNinja-1.0-OpenChat-7B - beowolx/MistralHermes-CodePro-7B-v1 model-index: - name: NinjaDolphin-7B results: - task: type: text-generation # Required. Example: automatic-speech-recognition dataset: type: openai_humaneval # Required. Example: common_voice. Use dataset id from https://hf.co/datasets name: HumanEval # Required. A pretty name for the dataset. Example: Common Voice (French) metrics: - type: pass@1 # Required. Example: wer. Use metric id from https://hf.co/metrics value: 52.4390243902439 # Required. Example: 20.90 name: pass@1 # Optional. Example: Test WER verified: false

NinjaDolphin-7B

NinjaDolphin-7B is a merge of the following models using:

Improving coding ability from FelixChao/WizardDolphin-7B.

HumanEval (uninstructed and no post-process)

Metric	Value
humaneval-python	52.4390243902439

🧩 Configuration

models:
  - model: FelixChao/WizardDolphin-7B
  - model: beowolx/CodeNinja-1.0-OpenChat-7B
    parameters:
      density: 0.53
      weight: 0.3
  - model: beowolx/MistralHermes-CodePro-7B-v1
    parameters:
      density: 0.53
      weight: 0.3
merge_method: dare_ties
base_model: FelixChao/WizardDolphin-7B
parameters:
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "FelixChao/NinjaDolphin-7B"
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"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	69.74
AI2 Reasoning Challenge (25-Shot)	65.61
HellaSwag (10-Shot)	85.35
MMLU (5-Shot)	64.43
TruthfulQA (0-shot)	54.94
Winogrande (5-shot)	80.27
GSM8k (5-shot)	67.85