TinyEnsemble-3x1.1B-TinyMoE

TinyEnsemble-3x1.1B-TinyMoE is a Mixture of Experts (MoE) made with the following models using LazyMergekit:

• cognitivecomputations/TinyDolphin-2.8-1.1b
• 78health/TinyLlama_1.1B-function-calling
• DaertML/TinyGauss-1.1B

🧩 Configuration

base_model: cognitivecomputations/TinyDolphin-2.8-1.1b
gate_mode: cheap_embed
dtype: bfloat16
experts:
  - source_model: cognitivecomputations/TinyDolphin-2.8-1.1b
    positive_prompts: ["write", "explain", "summarize", "how", "what", "acting"]
  - source_model: 78health/TinyLlama_1.1B-function-calling
    positive_prompts: ["code", "python", "javascript", "programming", "script", "run", "create"]
  - source_model: DaertML/TinyGauss-1.1B
    positive_prompts: ["count", "math", "algorithm", "crypto", "logic", "reason"]

💻 Usage

!pip install -qU transformers bitsandbytes accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "JoPmt/TinyEnsemble-3x1.1B-TinyMoE"

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"])