Instructions to use s-sahoo/duo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use s-sahoo/duo with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="s-sahoo/duo", trust_remote_code=True)# Load model directly from transformers import AutoModelForMaskedLM model = AutoModelForMaskedLM.from_pretrained("s-sahoo/duo", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use s-sahoo/duo with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "s-sahoo/duo" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "s-sahoo/duo", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/s-sahoo/duo
- SGLang
How to use s-sahoo/duo 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 "s-sahoo/duo" \ --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": "s-sahoo/duo", "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 "s-sahoo/duo" \ --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": "s-sahoo/duo", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use s-sahoo/duo with Docker Model Runner:
docker model run hf.co/s-sahoo/duo
Add pipeline tag and project page URL
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,12 +1,13 @@
|
|
| 1 |
---
|
| 2 |
-
library_name: transformers
|
| 3 |
-
license: apache-2.0
|
| 4 |
-
language:
|
| 5 |
-
- en
|
| 6 |
datasets:
|
| 7 |
- Skylion007/openwebtext
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
metrics:
|
| 9 |
- perplexity
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
## Using DUO
|
|
@@ -19,7 +20,7 @@ tokenizer = transformers.AutoTokenizer.from_pretrained('gpt2')
|
|
| 19 |
model = AutoModelForMaskedLM.from_pretrained('s-sahoo/duo')
|
| 20 |
```
|
| 21 |
For a hands-on example, check out this [Colab notebook](https://colab.research.google.com/drive/1Sf7R-dqdR6gq-H8nyZ9E3ZkyvqMTqcwq?usp=sharing).
|
| 22 |
-
For more information and implementation details, visit our github repository: [DUO](https://github.com/s-sahoo/duo)
|
| 23 |
|
| 24 |
## Model Details
|
| 25 |
The model, which has a context length of `1024` and is similar in size to GPT2-medium with approximately `130 million` non-embedding parameters,
|
|
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
datasets:
|
| 3 |
- Skylion007/openwebtext
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
library_name: transformers
|
| 7 |
+
license: apache-2.0
|
| 8 |
metrics:
|
| 9 |
- perplexity
|
| 10 |
+
pipeline_tag: text-generation
|
| 11 |
---
|
| 12 |
|
| 13 |
## Using DUO
|
|
|
|
| 20 |
model = AutoModelForMaskedLM.from_pretrained('s-sahoo/duo')
|
| 21 |
```
|
| 22 |
For a hands-on example, check out this [Colab notebook](https://colab.research.google.com/drive/1Sf7R-dqdR6gq-H8nyZ9E3ZkyvqMTqcwq?usp=sharing).
|
| 23 |
+
For more information and implementation details, visit our github repository: [DUO](https://github.com/s-sahoo/duo) and project page: [Project Page](https://s-sahoo.com/duo)
|
| 24 |
|
| 25 |
## Model Details
|
| 26 |
The model, which has a context length of `1024` and is similar in size to GPT2-medium with approximately `130 million` non-embedding parameters,
|