Instructions to use NousResearch/Nous-Capybara-7B-V1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NousResearch/Nous-Capybara-7B-V1 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="NousResearch/Nous-Capybara-7B-V1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NousResearch/Nous-Capybara-7B-V1")
model = AutoModelForCausalLM.from_pretrained("NousResearch/Nous-Capybara-7B-V1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NousResearch/Nous-Capybara-7B-V1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NousResearch/Nous-Capybara-7B-V1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/Nous-Capybara-7B-V1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NousResearch/Nous-Capybara-7B-V1

SGLang

How to use NousResearch/Nous-Capybara-7B-V1 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 "NousResearch/Nous-Capybara-7B-V1" \
    --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": "NousResearch/Nous-Capybara-7B-V1",
		"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 "NousResearch/Nous-Capybara-7B-V1" \
        --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": "NousResearch/Nous-Capybara-7B-V1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use NousResearch/Nous-Capybara-7B-V1 with Docker Model Runner:
```
docker model run hf.co/NousResearch/Nous-Capybara-7B-V1
```

LDJnr commited on Sep 25, 2023

Commit

8d487f8

1 Parent(s): 51a29ed

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -12,7 +12,7 @@ datasets:
 ## **Nous-Capybara-7B**
-A model created with the goal of a synergistic combination of different techniques used for SOTA models such as Evol-Instruct, Orca, Platypus, Lamini, FLASK and others, all into one lean holistically formed dataset and model. The seed instructions used for the start of synthesized conversations are largely based on highly acclaimed datasets like Airoboros, Know logic, EverythingLM, GPTeacher and even entirely new seed instructions derived from posts and discussions made on the website LessWrong, as well as being supplemented with certain multi-turn datasets like Dove(A successor to Puffin).
 Entirely contained within 20K training examples, mostly comprised of newly synthesized tokens never used for model training until now!

 ## **Nous-Capybara-7B**
+A model created with the goal of a synergistic combination of different techniques used for SOTA models such as Evol-Instruct, Orca, Vicuna, Lamini, FLASK and others, all into one lean holistically formed dataset and model. The seed instructions used for the start of synthesized conversations are largely based on highly acclaimed datasets like Airoboros, Know logic, EverythingLM, GPTeacher and even entirely new seed instructions derived from posts and discussions made on the website LessWrong, as well as being supplemented with certain multi-turn datasets like Dove(A successor to Puffin).
 Entirely contained within 20K training examples, mostly comprised of newly synthesized tokens never used for model training until now!