Instructions to use Harley-ml/Dillion-1.2M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Harley-ml/Dillion-1.2M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Harley-ml/Dillion-1.2M")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("Harley-ml/Dillion-1.2M") model = AutoModelForCausalLM.from_pretrained("Harley-ml/Dillion-1.2M") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use Harley-ml/Dillion-1.2M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Harley-ml/Dillion-1.2M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Harley-ml/Dillion-1.2M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Harley-ml/Dillion-1.2M
- SGLang
How to use Harley-ml/Dillion-1.2M 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 "Harley-ml/Dillion-1.2M" \ --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": "Harley-ml/Dillion-1.2M", "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 "Harley-ml/Dillion-1.2M" \ --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": "Harley-ml/Dillion-1.2M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Harley-ml/Dillion-1.2M with Docker Model Runner:
docker model run hf.co/Harley-ml/Dillion-1.2M
Update README.md
Browse files
README.md
CHANGED
|
@@ -108,7 +108,7 @@ We trained Dillion for 0.71 epochs on 14B (only saw ~9B) tokens of FineWeb-edu o
|
|
| 108 |
|
| 109 |
| Task | Metric | Dillion | SupraMini-v4-2M | Tenete-8M |
|
| 110 |
| -------- | --------------- | -------: | --------------: | ---------: |
|
| 111 |
-
| ARC Easy | acc_norm | 31.36% |
|
| 112 |
| BLiMP | acc | 62.94% | 60.70% | — |
|
| 113 |
| PiQA | acc_norm | 53.10% | 51.90% | 55.71% |
|
| 114 |
| SWAG | acc_norm | 30.36% | — | 32.97% |
|
|
|
|
| 108 |
|
| 109 |
| Task | Metric | Dillion | SupraMini-v4-2M | Tenete-8M |
|
| 110 |
| -------- | --------------- | -------: | --------------: | ---------: |
|
| 111 |
+
| ARC Easy | acc_norm | 31.36% | 31.50% | 31.94% |
|
| 112 |
| BLiMP | acc | 62.94% | 60.70% | — |
|
| 113 |
| PiQA | acc_norm | 53.10% | 51.90% | 55.71% |
|
| 114 |
| SWAG | acc_norm | 30.36% | — | 32.97% |
|