Qwen-2.5-base-agentic-2-7b

Qwen-2.5-base-agentic-2-7b is a merge of the following models using LazyMergekit:

🧩 Configuration

name: Qwen-2.5-base-agentic-2-7b
merge_method: sce
parameters:
  select_topk: 0.8
  normalize: true
dtype: float32
out_dtype: bfloat16
base_model: bunnycore/QwenMosaic-7B
tokenizer:
  source: base
models:
  - model: Jebadiah/Qwen-2.5-base-7b
  - model: Xiaojian9992024/Qwen2.5-Dyanka-7B-Preview
  - model: bunnycore/Blabbertron-1.2
  - model: allknowingroger/Marco-01-slerp3-7B
  - model: Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Jebadiah/Qwen-2.5-base-agentic-2-7b"
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"])