Instructions to use Doradus-AI/RnJ-1-Instruct-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Doradus-AI/RnJ-1-Instruct-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="Doradus-AI/RnJ-1-Instruct-FP8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Doradus-AI/RnJ-1-Instruct-FP8")
model = AutoModelForCausalLM.from_pretrained("Doradus-AI/RnJ-1-Instruct-FP8")
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 Doradus-AI/RnJ-1-Instruct-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Doradus-AI/RnJ-1-Instruct-FP8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Doradus-AI/RnJ-1-Instruct-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Doradus-AI/RnJ-1-Instruct-FP8

SGLang

How to use Doradus-AI/RnJ-1-Instruct-FP8 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 "Doradus-AI/RnJ-1-Instruct-FP8" \
    --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": "Doradus-AI/RnJ-1-Instruct-FP8",
		"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 "Doradus-AI/RnJ-1-Instruct-FP8" \
        --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": "Doradus-AI/RnJ-1-Instruct-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Doradus-AI/RnJ-1-Instruct-FP8 with Docker Model Runner:
```
docker model run hf.co/Doradus-AI/RnJ-1-Instruct-FP8
```

RnJ-1-Instruct-FP8 / docker /Dockerfile

doradus

Upload folder using huggingface_hub

6a70e5e verified 6 months ago

raw

history blame contribute delete

925 Bytes

	# RnJ-1-Instruct-FP8 Inference Server
	# Based on vLLM OpenAI-compatible server
	#
	# Build:
	# docker build -t rnj-1-instruct-fp8 .
	#
	# Run:
	# docker run --gpus '"device=0"' -p 8000:8000 rnj-1-instruct-fp8

	FROM vllm/vllm-openai:v0.12.0

	# Model will be downloaded from HuggingFace on first run
	ENV MODEL_NAME="Doradus/RnJ-1-Instruct-FP8"
	ENV MAX_MODEL_LEN="8192"
	ENV GPU_MEMORY_UTILIZATION="0.90"

	# vLLM settings
	ENV VLLM_ATTENTION_BACKEND="FLASHINFER"

	EXPOSE 8000

	ENTRYPOINT ["python", "-m", "vllm.entrypoints.openai.api_server"]

	CMD ["--model", "Doradus/RnJ-1-Instruct-FP8", \
	"--host", "0.0.0.0", \
	"--port", "8000", \
	"--tensor-parallel-size", "1", \
	"--max-model-len", "8192", \
	"--gpu-memory-utilization", "0.90", \
	"--dtype", "auto", \
	"--trust-remote-code", \
	"--served-model-name", "rnj-1-instruct-fp8", \
	"--enable-chunked-prefill", \
	"--max-num-seqs", "32"]