Instructions to use prefixsliding/1.5B-v84 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use prefixsliding/1.5B-v84 with Transformers:

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

pipe = pipeline("text-generation", model="prefixsliding/1.5B-v84")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("prefixsliding/1.5B-v84")
model = AutoModelForCausalLM.from_pretrained("prefixsliding/1.5B-v84", 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 prefixsliding/1.5B-v84 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prefixsliding/1.5B-v84"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "prefixsliding/1.5B-v84",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/prefixsliding/1.5B-v84

SGLang

How to use prefixsliding/1.5B-v84 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 "prefixsliding/1.5B-v84" \
    --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": "prefixsliding/1.5B-v84",
		"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 "prefixsliding/1.5B-v84" \
        --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": "prefixsliding/1.5B-v84",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use prefixsliding/1.5B-v84 with Docker Model Runner:
```
docker model run hf.co/prefixsliding/1.5B-v84
```

Muennighoff commited on Oct 4, 2025

Commit

239fa6f

verified ·

1 Parent(s): 8d4c5c9

Training in progress, step 200

Browse files

Files changed (3) hide show

README.md +2 -2
model.safetensors +1 -1
training_args.bin +1 -1

README.md CHANGED Viewed

@@ -3,8 +3,8 @@ library_name: transformers
 model_name: 1.5B-v84
 tags:
 - generated_from_trainer
-- grpo
 - trl
 licence: license
 ---
@@ -26,7 +26,7 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/muennighoff/s2/runs/zelv8q4r)
 This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).

 model_name: 1.5B-v84
 tags:
 - generated_from_trainer
 - trl
+- grpo
 licence: license
 ---
 ## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/muennighoff/s2/runs/mzdrm2rr)
 This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c40d45f56d4dcb40579d18d566a7e961221a4d148b10faa088fbbe027e313a96
 size 4063515640

 version https://git-lfs.github.com/spec/v1
+oid sha256:9781cdb3ae3fc3c1216e8493eda3db70540cec6aec934d3591ca1ac15f749f99
 size 4063515640

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1a9d12869ff3b1d26302933937ba1965f6df38135c17217be8c1f3b2780e48b
 size 8977

 version https://git-lfs.github.com/spec/v1
+oid sha256:558894e359492157a64c945450941236e61210dc01e2c8553ef101471f325bda
 size 8977