Instructions to use epfl-llm/meditron-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use epfl-llm/meditron-7b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="epfl-llm/meditron-7b")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("epfl-llm/meditron-7b") model = AutoModelForCausalLM.from_pretrained("epfl-llm/meditron-7b", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use epfl-llm/meditron-7b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "epfl-llm/meditron-7b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "epfl-llm/meditron-7b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/epfl-llm/meditron-7b
- SGLang
How to use epfl-llm/meditron-7b 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 "epfl-llm/meditron-7b" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "epfl-llm/meditron-7b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "epfl-llm/meditron-7b" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "epfl-llm/meditron-7b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use epfl-llm/meditron-7b with Docker Model Runner:
docker model run hf.co/epfl-llm/meditron-7b
Validation Dataset for Edge-Case Benchmarking (Post-Quantum / Rare Pathologies)
Hi everyone,
We have been benchmarking this model's performance on highly specific edge-case vectors. We noticed a structural data gap regarding advanced anomalies (post-quantum entropy drops / rare longitudinal EHR records like Susac & Wolfram syndromes).
To help the community validate these gaps, we published a 100k-row fully optimized, balanced synthetic evaluation dataset under restrictive non-commercial licensing:
• For Medical Clinical Evaluation: (https://huggingface.co/datasets/Synthis-Med-Labs/rare-diseases-synthetic-clinical-records-100k)
• For Post-Quantum Network Logs: (https://huggingface.co/datasets/Synthis-Med-Labs/quantum-cyber-attack-synthetic-logs-100k)
Feel free to pull the Parquet/CSV files to stress-test your weights and let us know your accuracy deltas in this thread.
Best,
Synthis Med Labs Team