Instructions to use neph1/sd-seer-tinyllama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use neph1/sd-seer-tinyllama with Transformers:

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

pipe = pipeline("text-generation", model="neph1/sd-seer-tinyllama")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("neph1/sd-seer-tinyllama")
model = AutoModelForCausalLM.from_pretrained("neph1/sd-seer-tinyllama")

llama-cpp-python

How to use neph1/sd-seer-tinyllama with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="neph1/sd-seer-tinyllama",
	filename="gguf/sd-seer-tinyllama-q8.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use neph1/sd-seer-tinyllama with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf neph1/sd-seer-tinyllama
# Run inference directly in the terminal:
llama-cli -hf neph1/sd-seer-tinyllama

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf neph1/sd-seer-tinyllama
# Run inference directly in the terminal:
llama-cli -hf neph1/sd-seer-tinyllama

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf neph1/sd-seer-tinyllama
# Run inference directly in the terminal:
./llama-cli -hf neph1/sd-seer-tinyllama

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf neph1/sd-seer-tinyllama
# Run inference directly in the terminal:
./build/bin/llama-cli -hf neph1/sd-seer-tinyllama

Use Docker

docker model run hf.co/neph1/sd-seer-tinyllama

LM Studio
Jan

vLLM

How to use neph1/sd-seer-tinyllama with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "neph1/sd-seer-tinyllama"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "neph1/sd-seer-tinyllama",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/neph1/sd-seer-tinyllama

SGLang

How to use neph1/sd-seer-tinyllama 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 "neph1/sd-seer-tinyllama" \
    --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": "neph1/sd-seer-tinyllama",
		"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 "neph1/sd-seer-tinyllama" \
        --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": "neph1/sd-seer-tinyllama",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use neph1/sd-seer-tinyllama with Ollama:
```
ollama run hf.co/neph1/sd-seer-tinyllama
```

Unsloth Studio new

How to use neph1/sd-seer-tinyllama with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for neph1/sd-seer-tinyllama to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for neph1/sd-seer-tinyllama to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for neph1/sd-seer-tinyllama to start chatting

Docker Model Runner
How to use neph1/sd-seer-tinyllama with Docker Model Runner:
```
docker model run hf.co/neph1/sd-seer-tinyllama
```

Lemonade

How to use neph1/sd-seer-tinyllama with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull neph1/sd-seer-tinyllama

Run and chat with the model

lemonade run user.sd-seer-tinyllama-{{QUANT_TAG}}

List all available models

lemonade list

Experimental and temperamental. Base idea:

https://huggingface.co/neph1/sd-seer-griffin-3b

Requires some retries to get a decent result. Maybe a prompt issue?

Example (cherry-picked):

Instruction:

A stunning and intricate statue of the Hindu deity Shiva is depicted in a cybertronic style, combining elements of Art Nouveau, Cyberpunk, and Gothic design. The statue features a feminine form with detailed cybernetic enhancements, showcasing the skill of renowned artists Alphonse Mucha, Ayami Kojima, Amano, Greg Hildebrandt, and Mark Brooks. This unique piece of art would make an impressive addition to any collection or exhibition. Response:

cybernetic shiva, intricate, elegant, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, art by alphonse mucha and ayami kojima and amano and greg hildebrandt and mark brooks

Prompt template:

<|system|>

This is a conversation between User and the helpful AI-bot Llama. Llama breaks down image descriptions into precise comma-separated single-word tags adding tags where they might increase quality. Llama responds only in comma-separated tags without asking questions.

<|user|>

A stunning and intricate statue of the Hindu deity Shiva is depicted in a cybertronic style, combining elements of Art Nouveau, Cyberpunk, and Gothic design. The statue features a feminine form with detailed cybernetic enhancements, showcasing the skill of renowned artists Alphonse Mucha, Ayami Kojima, Amano, Greg Hildebrandt, and Mark Brooks. This unique piece of art would make an impressive addition to any collection or exhibition.

<|assistant|>

...

Downloads last month: 50

Safetensors

Model size

1B params

Tensor type

F16

Datasets used to train neph1/sd-seer-tinyllama

Collection including neph1/sd-seer-tinyllama

sd-seer

Collection

Experimental models to convert natural language stable diffusion prompts to tags • 2 items • Updated Oct 20, 2024 • 1