Instructions to use PeymanHosseini/Hummingbird with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use PeymanHosseini/Hummingbird with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="PeymanHosseini/Hummingbird")# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("PeymanHosseini/Hummingbird", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use PeymanHosseini/Hummingbird with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "PeymanHosseini/Hummingbird" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "PeymanHosseini/Hummingbird", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/PeymanHosseini/Hummingbird
- SGLang
How to use PeymanHosseini/Hummingbird 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 "PeymanHosseini/Hummingbird" \ --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": "PeymanHosseini/Hummingbird", "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 "PeymanHosseini/Hummingbird" \ --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": "PeymanHosseini/Hummingbird", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use PeymanHosseini/Hummingbird with Docker Model Runner:
docker model run hf.co/PeymanHosseini/Hummingbird
Update README.md
Browse filesUpdates README (Fixes Grammer Errors)
README.md
CHANGED
|
@@ -16,14 +16,14 @@ This version of Hummingbird is only meant to demonstrate Efficient Attention for
|
|
| 16 |
|
| 17 |
## Model Details
|
| 18 |
|
| 19 |
-
The
|
| 20 |
|
| 21 |
| Parameter | size |
|
| 22 |
-
| -------------------
|
| 23 |
| # Transformer Blocks | 10 |
|
| 24 |
| Model Dimension | 3072 |
|
| 25 |
| # Heads | 1 |
|
| 26 |
|
| 27 |
|
| 28 |
-
The Attention Mechanism used is based on our newly proposed Efficient Attention from our paper, *You Need to Pay Better Attention: Rethinking the Mathematics of Attention Mechanism* ([arXiv:2403.01643](https://arxiv.org/abs/2403.01643)). We have chosen the number of
|
| 29 |
|
|
|
|
| 16 |
|
| 17 |
## Model Details
|
| 18 |
|
| 19 |
+
The model consists of 1.1 Billion parameters with the following specifications:
|
| 20 |
|
| 21 |
| Parameter | size |
|
| 22 |
+
| :------------------- | :--- |
|
| 23 |
| # Transformer Blocks | 10 |
|
| 24 |
| Model Dimension | 3072 |
|
| 25 |
| # Heads | 1 |
|
| 26 |
|
| 27 |
|
| 28 |
+
The Attention Mechanism used is based on our newly proposed Efficient Attention from our paper, *You Need to Pay Better Attention: Rethinking the Mathematics of Attention Mechanism* ([arXiv:2403.01643](https://arxiv.org/abs/2403.01643)). We have chosen the number of heads to be 1 as an interesting case study since all current LMs use multiple heads.
|
| 29 |
|