Instructions to use Epode/Prelude-1B-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Epode/Prelude-1B-Base with Transformers:

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

pipe = pipeline("text-generation", model="Epode/Prelude-1B-Base")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Epode/Prelude-1B-Base")
model = AutoModelForCausalLM.from_pretrained("Epode/Prelude-1B-Base", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Epode/Prelude-1B-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Epode/Prelude-1B-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Epode/Prelude-1B-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Epode/Prelude-1B-Base

SGLang

How to use Epode/Prelude-1B-Base 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 "Epode/Prelude-1B-Base" \
    --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": "Epode/Prelude-1B-Base",
		"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 "Epode/Prelude-1B-Base" \
        --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": "Epode/Prelude-1B-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Epode/Prelude-1B-Base with Docker Model Runner:
```
docker model run hf.co/Epode/Prelude-1B-Base
```

Prelude-1B-Base

File size: 1,179 Bytes

7cd1da6

{
  "checkpoint": "/home/hhai/pretrain/checkpoints/full/final/lit_model.pth",
  "evaluation": {
    "batch_size": "auto:4",
    "dtype": "bfloat16",
    "lm_eval_version": "0.4.9",
    "seed": 42
  },
  "final_pretraining": {
    "tokens": 19999703040,
    "validation_loss": 2.4247145652770996,
    "validation_perplexity": 11.299003601074219
  },
  "gsm8k_5shot": {
    "exact_match_flexible_extract": 0.016679302501895376,
    "exact_match_strict_match": 0.00530705079605762
  },
  "mmlu_5shot": {
    "accuracy": 0.2530978493092152,
    "accuracy_stderr": 0.0036649761750357874,
    "context_limit": 2048,
    "left_truncation_observed": true
  },
  "raw_results_in_git": false,
  "selected_metric_core_mean": 0.4785265940898973,
  "zero_shot_core": {
    "arc_challenge_acc_norm": 0.3361774744027304,
    "arc_easy_acc_norm": 0.6414141414141414,
    "hellaswag_acc_norm": 0.45130452101175067,
    "lambada_openai_accuracy": 0.4626431205123229,
    "lambada_openai_perplexity": 13.241914520713838,
    "openbookqa_acc_norm": 0.35,
    "piqa_acc_norm": 0.6958650707290533,
    "truthfulqa_mc2_accuracy": 0.3691036101266861,
    "winogrande_accuracy": 0.5217048145224941
  }
}