Instructions to use abacusai/MetaMath-bagel-34b-v0.2-c1500 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use abacusai/MetaMath-bagel-34b-v0.2-c1500 with Transformers:

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

pipe = pipeline("text-generation", model="abacusai/MetaMath-bagel-34b-v0.2-c1500")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

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

tokenizer = AutoTokenizer.from_pretrained("abacusai/MetaMath-bagel-34b-v0.2-c1500")
model = AutoModelForCausalLM.from_pretrained("abacusai/MetaMath-bagel-34b-v0.2-c1500", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use abacusai/MetaMath-bagel-34b-v0.2-c1500 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "abacusai/MetaMath-bagel-34b-v0.2-c1500"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "abacusai/MetaMath-bagel-34b-v0.2-c1500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/abacusai/MetaMath-bagel-34b-v0.2-c1500

SGLang

How to use abacusai/MetaMath-bagel-34b-v0.2-c1500 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 "abacusai/MetaMath-bagel-34b-v0.2-c1500" \
    --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": "abacusai/MetaMath-bagel-34b-v0.2-c1500",
		"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 "abacusai/MetaMath-bagel-34b-v0.2-c1500" \
        --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": "abacusai/MetaMath-bagel-34b-v0.2-c1500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use abacusai/MetaMath-bagel-34b-v0.2-c1500 with Docker Model Runner:
```
docker model run hf.co/abacusai/MetaMath-bagel-34b-v0.2-c1500
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Finetune of the pre-DPO Bagel model (https://huggingface.co/jondurbin/bagel-34b-v0.2) on the MetamathFewshot (https://huggingface.co/datasets/abacusai/MetaMathFewshot) dataset

Evaluation Results

Average	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K

For comparison the GSM8K score for the original metamath/MetaMath-Mistral-7B was 46.17 and average score was 69.7

Downloads last month: 71

Safetensors

Model size

34B params

Tensor type

F16

Model tree for abacusai/MetaMath-bagel-34b-v0.2-c1500

Quantizations

1 model

abacusai
/

MetaMath-bagel-34b-v0.2-c1500

Evaluation Results

Model tree for abacusai/MetaMath-bagel-34b-v0.2-c1500

Dataset used to train abacusai/MetaMath-bagel-34b-v0.2-c1500