Fine-tuned models released with a research project, one folder per model. Each folder's name is an opaque id; the project's code repository maps it to its training run (models.csv) and names its base model. Each model keeps its base model's licence (Qwen2.5-3B: Qwen research licence; Qwen2.5-7B/14B/32B, Ministral-3: Apache-2.0; Llama-3.1: Llama 3.1 licence).
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("janniskirschner/mtds", subfolder="<id>")
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support