Instructions to use SultanR/SmolTulu-1.7b-Reinforced with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SultanR/SmolTulu-1.7b-Reinforced with Transformers:

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

pipe = pipeline("text-generation", model="SultanR/SmolTulu-1.7b-Reinforced")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SultanR/SmolTulu-1.7b-Reinforced")
model = AutoModelForCausalLM.from_pretrained("SultanR/SmolTulu-1.7b-Reinforced", 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 SultanR/SmolTulu-1.7b-Reinforced with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SultanR/SmolTulu-1.7b-Reinforced"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SultanR/SmolTulu-1.7b-Reinforced",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SultanR/SmolTulu-1.7b-Reinforced

SGLang

How to use SultanR/SmolTulu-1.7b-Reinforced 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 "SultanR/SmolTulu-1.7b-Reinforced" \
    --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": "SultanR/SmolTulu-1.7b-Reinforced",
		"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 "SultanR/SmolTulu-1.7b-Reinforced" \
        --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": "SultanR/SmolTulu-1.7b-Reinforced",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SultanR/SmolTulu-1.7b-Reinforced with Docker Model Runner:
```
docker model run hf.co/SultanR/SmolTulu-1.7b-Reinforced
```

SultanR commited on Dec 17, 2024

Commit

2116ae7

verified ·

1 Parent(s): f999641

Update README.md

Browse files

Files changed (1) hide show

README.md +19 -0

README.md CHANGED Viewed

@@ -29,6 +29,7 @@ pipeline_tag: text-generation
 SmolTulu-1.7b-Reinforced is the reinforcement learning with verifiable rewards (RLVR) version of [SmolTulu-1.7b-Instruct](https://huggingface.co/SultanR/SmolTulu-1.7b-Instruct), which leverages [AllenAI's Tulu 3 post-training pipeline](https://arxiv.org/abs/2411.15124)
 This model scores the highest current score in both IFEval and GSM8k while maintaining the extremely low contamination levels in Tulu 3 and SmolLM2! I've listed the datasets used to do both the RLVR stage, which is the same one mentioned used in the Tulu 3 paper.
 ## Evaluation
 I ran these evaluations using [SmolLM2's evaluation code](https://github.com/huggingface/smollm/tree/main/evaluation) for a more fair comparison.
@@ -44,6 +45,24 @@ I ran these evaluations using [SmolLM2's evaluation code](https://github.com/hug
 | MMLU-Pro (MCF) | 17.4 | 17.3 | 19.3 | 12.7 | **24.2** | 11.7 |
 | PIQA | 72.2 | 72.1 | **74.4** | 72.3 | 73.2 | 71.6 |
 ## Usage
 Just like any Huggingface model, just run it using the transformers library:

 SmolTulu-1.7b-Reinforced is the reinforcement learning with verifiable rewards (RLVR) version of [SmolTulu-1.7b-Instruct](https://huggingface.co/SultanR/SmolTulu-1.7b-Instruct), which leverages [AllenAI's Tulu 3 post-training pipeline](https://arxiv.org/abs/2411.15124)
 This model scores the highest current score in both IFEval and GSM8k while maintaining the extremely low contamination levels in Tulu 3 and SmolLM2! I've listed the datasets used to do both the RLVR stage, which is the same one mentioned used in the Tulu 3 paper.
 ## Evaluation
 I ran these evaluations using [SmolLM2's evaluation code](https://github.com/huggingface/smollm/tree/main/evaluation) for a more fair comparison.
 | MMLU-Pro (MCF) | 17.4 | 17.3 | 19.3 | 12.7 | **24.2** | 11.7 |
 | PIQA | 72.2 | 72.1 | **74.4** | 72.3 | 73.2 | 71.6 |
+## Training Details
+The reinforced model used PPO with verifiable rewards:
+- Base model: SmolTulu-1.7b-Instruct
+- Learning rate: 3e-6
+- Total training episodes: 10M
+- PPO KL penalty coefficient (beta): 0.05
+- Maximum sequence/prompt length: 2048 tokens
+- Response length: 2048 tokens
+- Rollout batch size: 32
+- Minibatch size: 32
+- Temperature: 1.0
+- Penalty reward: -10.0 for incomplete generations
+- DeepSpeed Stage 3 optimization
+- Gradient checkpointing enabled
+- Training data: RLVR-GSM-MATH-IF-Mixed-Constraints
+- Reward model multiplier: 0.0 (pure verifiable rewards)
 ## Usage
 Just like any Huggingface model, just run it using the transformers library: