FMT-v1

FMT-v1 is a merge of the following models using LazyMergekit:

• FuseAI/FuseChat-7B-VaRM
• MediaTek-Research/Breeze-7B-Instruct-v1_0
• tokyotech-llm/Swallow-MS-7b-v0.1

🧩 Configuration

models:
  - model: FuseAI/FuseChat-7B-VaRM
    parameters:
      weight: 1.0
  - model: MediaTek-Research/Breeze-7B-Instruct-v1_0
    parameters:
      weight: 0.75
  - model: tokyotech-llm/Swallow-MS-7b-v0.1
    parameters:
      weight: 0.55
merge_method: task_arithmetic
base_model: FuseAI/FuseChat-7B-VaRM
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Virt-io/FMT-v1"
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"])