Instructions to use yolay/RAIF-DeepScaleR-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yolay/RAIF-DeepScaleR-1.5B with Transformers:

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

pipe = pipeline("text-generation", model="yolay/RAIF-DeepScaleR-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yolay/RAIF-DeepScaleR-1.5B")
model = AutoModelForCausalLM.from_pretrained("yolay/RAIF-DeepScaleR-1.5B")
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

vLLM

How to use yolay/RAIF-DeepScaleR-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yolay/RAIF-DeepScaleR-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yolay/RAIF-DeepScaleR-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yolay/RAIF-DeepScaleR-1.5B

SGLang

How to use yolay/RAIF-DeepScaleR-1.5B 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 "yolay/RAIF-DeepScaleR-1.5B" \
    --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": "yolay/RAIF-DeepScaleR-1.5B",
		"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 "yolay/RAIF-DeepScaleR-1.5B" \
        --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": "yolay/RAIF-DeepScaleR-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yolay/RAIF-DeepScaleR-1.5B with Docker Model Runner:
```
docker model run hf.co/yolay/RAIF-DeepScaleR-1.5B
```

Add library name, pipeline tag, and Github link

by nielsr HF Staff - opened Jun 4, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

-2

Files changed (1) hide show

README.md +5 -2

README.md CHANGED Viewed

@@ -1,8 +1,11 @@
 ---
-license: apache-2.0
 datasets:
 - yolay/RAIF-ComplexInstruction-DeepSeek
 ---
 This model belongs to the official implementation of the paper "Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models".
 Existing large language models (LLMs) face challenges of following complex instructions, especially when multiple constraints are present and organized in paralleling, chaining, and branching structures. One intuitive solution, namely chain-of-thought (CoT), is expected to universally improve capabilities of LLMs. However, we find that the vanilla CoT exerts a negative impact on performance due to its superficial reasoning pattern of simply paraphrasing the instructions. It fails to peel back the compositions of constraints for identifying their relationship across hierarchies of types and dimensions.
@@ -45,7 +48,7 @@ The model DeepScaleR-1.5B is our optimized model for its advanced instruction-fo
 | DeepSeek-Qwen7B        | SFT      | 67.09  | 69.10 | 58.66    | 58.42        | 55.60    | 65.96        | 79.15      | 64.85 (-0.88%)  |
 | DeepSeek-Qwen7B        | Ours     | 71.35  | 71.40 | 58.67    | 62.04        | 59.65    | 59.38        | 82.00      | 66.35 (+0.62%) |
 🎓 If you find this work useful, please consider the following citation:
 ```

 ---
 datasets:
 - yolay/RAIF-ComplexInstruction-DeepSeek
+license: apache-2.0
+library_name: transformers
+pipeline_tag: text-generation
 ---
 This model belongs to the official implementation of the paper "Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models".
 Existing large language models (LLMs) face challenges of following complex instructions, especially when multiple constraints are present and organized in paralleling, chaining, and branching structures. One intuitive solution, namely chain-of-thought (CoT), is expected to universally improve capabilities of LLMs. However, we find that the vanilla CoT exerts a negative impact on performance due to its superficial reasoning pattern of simply paraphrasing the instructions. It fails to peel back the compositions of constraints for identifying their relationship across hierarchies of types and dimensions.
 | DeepSeek-Qwen7B        | SFT      | 67.09  | 69.10 | 58.66    | 58.42        | 55.60    | 65.96        | 79.15      | 64.85 (-0.88%)  |
 | DeepSeek-Qwen7B        | Ours     | 71.35  | 71.40 | 58.67    | 62.04        | 59.65    | 59.38        | 82.00      | 66.35 (+0.62%) |
+Code: https://github.com/yuleiqin/RAIF
 🎓 If you find this work useful, please consider the following citation:
 ```