Instructions to use tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2") model = AutoModelForMultimodalLM.from_pretrained("tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2
- SGLang
How to use tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2 with Docker Model Runner:
docker model run hf.co/tepirale/Ornith-Agents-A1-3.6-35B-A3B-task_arithmetic_v2
What is difference between v1 and v2?
Is there any difference in the model merge method?
The first version of MergeKit required some libraries, which I didn't realize, but the model merge went smoothly and without issue.
I tested the first version, and it worked perfectly in VLLM.
For the second version, I installed the required libraries and performed the merge without any errors.
Is the DARE TIES merge method more advanced than the Task Arithmetical method, or does it have its own pros and cons?
Dare-ties might be superior, but I plan to merge it with other Mergekit methods and ultimately merge a variety of models with date-ties.
I'm adding this:
With task_arithmetic_v2, I've noticed that when the model makes a mistake, it recognizes it and corrects it (generating more tokens). However, with dare_ties, it can very rarely make a mistake, but it doesn't correct it. I want to see if merging them afterward captures this reasoning.
Wow that's interesting! Thank you for your contribution.
I'll look forward to your model release!