Instructions to use MarsupialAI/Dumbstral-169B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MarsupialAI/Dumbstral-169B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MarsupialAI/Dumbstral-169B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("MarsupialAI/Dumbstral-169B") model = AutoModelForCausalLM.from_pretrained("MarsupialAI/Dumbstral-169B") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use MarsupialAI/Dumbstral-169B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MarsupialAI/Dumbstral-169B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MarsupialAI/Dumbstral-169B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/MarsupialAI/Dumbstral-169B
- SGLang
How to use MarsupialAI/Dumbstral-169B 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 "MarsupialAI/Dumbstral-169B" \ --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": "MarsupialAI/Dumbstral-169B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'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 "MarsupialAI/Dumbstral-169B" \ --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": "MarsupialAI/Dumbstral-169B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use MarsupialAI/Dumbstral-169B with Docker Model Runner:
docker model run hf.co/MarsupialAI/Dumbstral-169B
Wow!!!
Hi! The model turned out great! It was my two best and most fun weekends. the model has a good consistency in the logic of actions and is really creative! I loved putting my character in awkward situations and dialogues so much) His answers, and especially mental dialogues with himself, are certainly something! He self-destructed himself. The model describes the situation, objects and creatures in detail very beautifully. But the only thing that the model is very adaptable to previous messages: that is, it turns from a regular chat into something literary... Heh, apparently I just don't like this) This model imitates the character of the character very well. and I would be very glad to see this model in aihorde again at least once!
Sincerely, from Alexandria.
Hi! The model turned out great! It was my two best and most fun weekends. the model has a good consistency in the logic of actions and is really creative! I loved putting my character in awkward situations and dialogues so much) His answers, and especially mental dialogues with himself, are certainly something! He self-destructed himself. The model describes the situation, objects and creatures in detail very beautifully. But the only thing that the model is very adaptable to previous messages: that is, it turns from a regular chat into something literary... Heh, apparently I just don't like this) This model imitates the character of the character very well. and I would be very glad to see this model in aihorde again at least once!
Sincerely, from Alexandria.
Can you share your sampler settings? Thanks.