You can deploy this model and start earning money today!

RichardErkhov/
shadowml_-_Beyonder-4x7B-v2-gguf

Name: shadowml_-_Beyonder-4x7B-v2-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Beyonder-4x7B-v2 - GGUF

Model creator: https://huggingface.co/shadowml/
Original model: https://huggingface.co/shadowml/Beyonder-4x7B-v2/

Name	Quant method	Size
Beyonder-4x7B-v2.Q2_K.gguf	Q2_K	8.24GB
Beyonder-4x7B-v2.IQ3_XS.gguf	IQ3_XS	9.21GB
Beyonder-4x7B-v2.IQ3_S.gguf	IQ3_S	9.73GB
Beyonder-4x7B-v2.Q3_K_S.gguf	Q3_K_S	9.72GB
Beyonder-4x7B-v2.IQ3_M.gguf	IQ3_M	9.92GB
Beyonder-4x7B-v2.Q3_K.gguf	Q3_K	10.79GB
Beyonder-4x7B-v2.Q3_K_M.gguf	Q3_K_M	10.79GB
Beyonder-4x7B-v2.Q3_K_L.gguf	Q3_K_L	11.68GB
Beyonder-4x7B-v2.IQ4_XS.gguf	IQ4_XS	12.15GB
Beyonder-4x7B-v2.Q4_0.gguf	Q4_0	12.69GB
Beyonder-4x7B-v2.IQ4_NL.gguf	IQ4_NL	4.29GB
Beyonder-4x7B-v2.Q4_K_S.gguf	Q4_K_S	12.8GB
Beyonder-4x7B-v2.Q4_K.gguf	Q4_K	13.61GB
Beyonder-4x7B-v2.Q4_K_M.gguf	Q4_K_M	13.61GB
Beyonder-4x7B-v2.Q4_1.gguf	Q4_1	14.09GB
Beyonder-4x7B-v2.Q5_0.gguf	Q5_0	15.48GB
Beyonder-4x7B-v2.Q5_K_S.gguf	Q5_K_S	15.48GB
Beyonder-4x7B-v2.Q5_K.gguf	Q5_K	15.96GB
Beyonder-4x7B-v2.Q5_K_M.gguf	Q5_K_M	15.96GB
Beyonder-4x7B-v2.Q5_1.gguf	Q5_1	16.88GB
Beyonder-4x7B-v2.Q6_K.gguf	Q6_K	18.46GB
Beyonder-4x7B-v2.Q8_0.gguf	Q8_0	23.9GB

Original model description:

license: apache-2.0 tags:

moe
mergekit
merge

Beyonder-4x7B-v2

This model is a Mixure of Experts (MoE) made with mergekit (mixtral branch). It uses the following base models:

🧩 Configuration

yaml

base_model: mlabonne/Marcoro14-7B-slerp
experts:
  - source_model: openchat/openchat-3.5-1210
    positive_prompts:
    - "chat"
    - "assistant"
    - "tell me"
    - "explain"
  - source_model: beowolx/CodeNinja-1.0-OpenChat-7B
    positive_prompts:
    - "code"
    - "python"
    - "javascript"
    - "programming"
    - "algorithm"
  - source_model: maywell/PiVoT-0.1-Starling-LM-RP
    positive_prompts:
    - "storywriting"
    - "write"
    - "scene"
    - "story"
    - "character"
  - source_model: WizardLM/WizardMath-7B-V1.1
    positive_prompts:
    - "reason"
    - "math"
    - "mathematics"
    - "solve"
    - "count"

💻 Usage

python

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Beyonder-4x7B-v2"

tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)

messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
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"])

RichardErkhov/shadowml_-_Beyonder-4x7B-v2-gguf

Original model description:

Beyonder-4x7B-v2

🧩 Configuration

💻 Usage

RichardErkhov/
shadowml_-_Beyonder-4x7B-v2-gguf