Instructions to use DSACXZ12EDSA/best-trained-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DSACXZ12EDSA/best-trained-model with Transformers:

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

pipe = pipeline("text-generation", model="DSACXZ12EDSA/best-trained-model")

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("DSACXZ12EDSA/best-trained-model")
model = AutoModel.from_pretrained("DSACXZ12EDSA/best-trained-model", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use DSACXZ12EDSA/best-trained-model with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DSACXZ12EDSA/best-trained-model"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DSACXZ12EDSA/best-trained-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DSACXZ12EDSA/best-trained-model

SGLang

How to use DSACXZ12EDSA/best-trained-model 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 "DSACXZ12EDSA/best-trained-model" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DSACXZ12EDSA/best-trained-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "DSACXZ12EDSA/best-trained-model" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DSACXZ12EDSA/best-trained-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use DSACXZ12EDSA/best-trained-model with Docker Model Runner:
```
docker model run hf.co/DSACXZ12EDSA/best-trained-model
```

MyAwesomeModel (Best Checkpoint: step_1000)

This is the highest-performing checkpoint (step 1000) of MyAwesomeModel, selected for maximum evaluation accuracy across all benchmarks.

Detailed Evaluation Results (all scores to 3 decimal places)

Benchmark Category	Score
Math Reasoning	0.875
Logical Reasoning	0.862
Common Sense	0.793
Reading Comprehension	0.764
Question Answering	0.721
Text Classification	0.857
Sentiment Analysis	0.832
Code Generation	0.789
Creative Writing	0.715
Dialogue Generation	0.763
Summarization	0.812
Translation	0.845
Knowledge Retrieval	0.758
Instruction Following	0.827
Safety Evaluation	0.796

Overall Weighted Score

Using the official weighted averaging scheme (with higher weight for reasoning/code tasks), the overall weighted performance score for this best checkpoint is 0.812 (3 decimal places).

This checkpoint demonstrates the model's peak performance, including the 87.5% accuracy on the AIME 2025 test set noted in the project documentation.

Downloads last month: 244