Instructions to use jaigouk/MetaMath-Cybertron-Starling-Ruby with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jaigouk/MetaMath-Cybertron-Starling-Ruby with Transformers:

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

pipe = pipeline("text-generation", model="jaigouk/MetaMath-Cybertron-Starling-Ruby")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jaigouk/MetaMath-Cybertron-Starling-Ruby")
model = AutoModelForCausalLM.from_pretrained("jaigouk/MetaMath-Cybertron-Starling-Ruby")
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 jaigouk/MetaMath-Cybertron-Starling-Ruby with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/jaigouk/MetaMath-Cybertron-Starling-Ruby

SGLang

How to use jaigouk/MetaMath-Cybertron-Starling-Ruby 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 "jaigouk/MetaMath-Cybertron-Starling-Ruby" \
    --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": "jaigouk/MetaMath-Cybertron-Starling-Ruby",
		"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 "jaigouk/MetaMath-Cybertron-Starling-Ruby" \
        --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": "jaigouk/MetaMath-Cybertron-Starling-Ruby",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jaigouk/MetaMath-Cybertron-Starling-Ruby with Docker Model Runner:
```
docker model run hf.co/jaigouk/MetaMath-Cybertron-Starling-Ruby
```

MetaMath-Cybertron-Starling-Ruby / README.md

jaigouk

Model save

b040121 over 2 years ago

preview code

raw

history blame contribute delete

1.52 kB

	---
	license: cc-by-nc-4.0
	base_model: Q-bert/MetaMath-Cybertron-Starling
	tags:
	- generated_from_trainer
	model-index:
	- name: MetaMath-Cybertron-Starling-Ruby
	results: []
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	# MetaMath-Cybertron-Starling-Ruby

	This model is a fine-tuned version of [Q-bert/MetaMath-Cybertron-Starling](https://huggingface.co/Q-bert/MetaMath-Cybertron-Starling) on an unknown dataset.
	It achieves the following results on the evaluation set:
	- Loss: 1.0319

	## Model description

	More information needed

	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 2e-05
	- train_batch_size: 16
	- eval_batch_size: 16
	- seed: 42
	- gradient_accumulation_steps: 4
	- total_train_batch_size: 64
	- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
	- lr_scheduler_type: constant
	- lr_scheduler_warmup_steps: 1
	- num_epochs: 1

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:-----:\|:----:\|:---------------:\|
	\| 1.0913 \| 0.39 \| 50 \| 1.0976 \|
	\| 1.0399 \| 0.78 \| 100 \| 1.0319 \|


	### Framework versions

	- Transformers 4.35.0
	- Pytorch 2.1.0+cu118
	- Datasets 2.14.5
	- Tokenizers 0.14.1