Instructions to use zjr2000/SPES-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zjr2000/SPES-2B with Transformers:

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

pipe = pipeline("text-generation", model="zjr2000/SPES-2B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zjr2000/SPES-2B")
model = AutoModelForCausalLM.from_pretrained("zjr2000/SPES-2B")
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 zjr2000/SPES-2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zjr2000/SPES-2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zjr2000/SPES-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zjr2000/SPES-2B

SGLang

How to use zjr2000/SPES-2B 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 "zjr2000/SPES-2B" \
    --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": "zjr2000/SPES-2B",
		"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 "zjr2000/SPES-2B" \
        --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": "zjr2000/SPES-2B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zjr2000/SPES-2B with Docker Model Runner:
```
docker model run hf.co/zjr2000/SPES-2B
```

zjr2000 commited on Mar 10

Commit

cd4580d

verified ·

1 Parent(s): 7f3a591

Update README.md

Browse files

Files changed (1) hide show

README.md +0 -7

README.md CHANGED Viewed

@@ -19,10 +19,6 @@ SPES-2B is a pretrained language model released as part of paper:
 **Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm**
-## Overview
-SPES-2B is a research checkpoint for studying decentralized and memory-efficient pretraining of Mixture-of-Experts (MoE) language models. The model is trained under the SPES framework, which enables distributed training across GPU nodes with lightweight expert synchronization.
 ## Model Details
 - **Model name:** SPES-2B
@@ -44,9 +40,6 @@ This model is intended for:
 - research on MoE training and synchronization
 - experimentation and evaluation of pretrained language models
-## Notes
-This is a research model. Performance may vary depending on evaluation setup, prompting format, and downstream tasks.
 ## Citation

 **Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm**
 ## Model Details
 - **Model name:** SPES-2B
 - research on MoE training and synchronization
 - experimentation and evaluation of pretrained language models
 ## Citation