Instructions to use MathAndMagic/polaris-small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MathAndMagic/polaris-small with Transformers:

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

pipe = pipeline("text-generation", model="MathAndMagic/polaris-small", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MathAndMagic/polaris-small")
model = AutoModelForCausalLM.from_pretrained("MathAndMagic/polaris-small", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MathAndMagic/polaris-small with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MathAndMagic/polaris-small"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MathAndMagic/polaris-small",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MathAndMagic/polaris-small

SGLang

How to use MathAndMagic/polaris-small 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 "MathAndMagic/polaris-small" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MathAndMagic/polaris-small",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "MathAndMagic/polaris-small" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MathAndMagic/polaris-small",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MathAndMagic/polaris-small with Docker Model Runner:
```
docker model run hf.co/MathAndMagic/polaris-small
```

ri-nat commited on Jan 28, 2024

Commit

9b58ac0

verified ·

1 Parent(s): ee7c793

Update README.md

Browse files

Files changed (1) hide show

README.md +8 -4

README.md CHANGED Viewed

@@ -25,7 +25,11 @@ This model was developed by StarfleetAI with the aim of it being used for autono
 ## Current Features
-As of version `0.1` the model is:
 - Able to call functions
 - Not bad at coding tasks
@@ -92,10 +96,10 @@ For now, the model is only capable of calling one function at a time.
 ## Function Call Response
-The model expects the function call response to be provided right after the function call request from a role `fn`:
 ```
-<|im_start|>fn
 {"result": "87cc47fbc865a290d7c7de4be3c893175c51a566b3"}<|im_end|>
 ```
@@ -122,7 +126,7 @@ You are a helpful assistant.<|im_end|>
 Generate a password, 42 characters long<|im_end|>
 <|im_start|>assistant
 <|fn_start|>{"name": "generate_password", "arguments": {"length": 42}}<|fn_end|><|im_end|>
-<|im_start|>fn
 {"result": "87cc47fbc865a290d7c7de4be3c893175c51a566b3"}<|im_end|>
 <|im_start|>assistant
 Here is your random password: 87cc47fbc865a290d7c7de4be3c893175c51a566b3. Please make sure to save it in a secure place.<|im_end|>

 ## Current Features
+### `v0.1.1`
+- Function call response role changed to `tool` to comply with OpenAI
+### `v0.1`
 - Able to call functions
 - Not bad at coding tasks
 ## Function Call Response
+The model expects the function call response to be provided right after the function call request from a role `tool`:
 ```
+<|im_start|>tool
 {"result": "87cc47fbc865a290d7c7de4be3c893175c51a566b3"}<|im_end|>
 ```
 Generate a password, 42 characters long<|im_end|>
 <|im_start|>assistant
 <|fn_start|>{"name": "generate_password", "arguments": {"length": 42}}<|fn_end|><|im_end|>
+<|im_start|>tool
 {"result": "87cc47fbc865a290d7c7de4be3c893175c51a566b3"}<|im_end|>
 <|im_start|>assistant
 Here is your random password: 87cc47fbc865a290d7c7de4be3c893175c51a566b3. Please make sure to save it in a secure place.<|im_end|>