Instructions to use bardsai/jaskier-7b-dpo-v6.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bardsai/jaskier-7b-dpo-v6.1 with Transformers:

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

pipe = pipeline("text-generation", model="bardsai/jaskier-7b-dpo-v6.1")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("bardsai/jaskier-7b-dpo-v6.1")
model = AutoModelForCausalLM.from_pretrained("bardsai/jaskier-7b-dpo-v6.1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use bardsai/jaskier-7b-dpo-v6.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bardsai/jaskier-7b-dpo-v6.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bardsai/jaskier-7b-dpo-v6.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bardsai/jaskier-7b-dpo-v6.1

SGLang

How to use bardsai/jaskier-7b-dpo-v6.1 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 "bardsai/jaskier-7b-dpo-v6.1" \
    --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": "bardsai/jaskier-7b-dpo-v6.1",
		"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 "bardsai/jaskier-7b-dpo-v6.1" \
        --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": "bardsai/jaskier-7b-dpo-v6.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bardsai/jaskier-7b-dpo-v6.1 with Docker Model Runner:
```
docker model run hf.co/bardsai/jaskier-7b-dpo-v6.1
```

Karols commited on Feb 20, 2024

Commit

e050aee

verified ·

1 Parent(s): aa3528c

Update README.md

Browse files

Files changed (1) hide show

README.md +8 -25

README.md CHANGED Viewed

@@ -20,38 +20,21 @@ Model based on `bardsai/jaskier-7b-dpo-v5.6` (downstream version of Mistral7B) f
 You can use this model directly with a Hugging Face pipeline:
 ```python
-import ...
-base_model_name = "bardsai/jaskier-7b-dpo-v5.6"
-p = pipeline("conversational", model=base_model_name, torch_dtype=torch.float16, device_map="auto")
-messages = [{'messages':[{
-    "role": "user",
-    "content": "How much is 2+2"
-}]},
-{'messages':[{
-    "role": "user",
-    "content": "Tell me where the nearest water park is?"
-}]},
-]
-def conversations():
-    for x in tqdm(messages):
-        yield Conversation(x['messages'])
-generate_kwargs = {
-    "do_sample": True,
-    "temperature": 0.7,
-    "max_new_tokens": 1024,
-}
-model_outputs = [out[-1]['content'] for out in p(conversations(), **generate_kwargs)]
-print(model_outputs)
 ```
 ## Output
->["The answer to the arithmetic operation of adding two to two is 4. In mathematical notation, it's written as 2 + 2 = 4.", "As an AI, I don't have physical proximity or real-time environment awareness. To provide location-based information, I need to be integrated with a specific location-tracking system or connected to a comprehensive database. Please check for local business listings, travel guides, or conduct an online search to find the nearest water park according to your current location."]
 ## Changelog

 You can use this model directly with a Hugging Face pipeline:
 ```python
+from transformers import pipeline, Conversation
+import torch
+base_model_name = "bardsai/jaskier-7b-dpo-v6.1"
+chatbot = pipeline("conversational", model=base_model_name, torch_dtype=torch.float16, device_map="auto")
+conversation = Conversation("Can Poland into space?")
+conversation = chatbot(conversation)
+print(conversation.messages[-1]["content"])
 ```
 ## Output
+"Poland, as a nation, doesn't physically travel to space. However, Poland has contributed to the field of space exploration through its scientists, engineers, and collaborations with international space agencies. The Polish Space Agency, established in 2016, aims to promote and coordinate the country's space activities."
 ## Changelog