Instructions to use KBlueLeaf/TIPO-100M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KBlueLeaf/TIPO-100M with Transformers:

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

pipe = pipeline("text-generation", model="KBlueLeaf/TIPO-100M")

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

tokenizer = AutoTokenizer.from_pretrained("KBlueLeaf/TIPO-100M")
model = AutoModelForCausalLM.from_pretrained("KBlueLeaf/TIPO-100M", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use KBlueLeaf/TIPO-100M with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf KBlueLeaf/TIPO-100M:F16
# Run inference directly in the terminal:
llama cli -hf KBlueLeaf/TIPO-100M:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf KBlueLeaf/TIPO-100M:F16
# Run inference directly in the terminal:
llama cli -hf KBlueLeaf/TIPO-100M:F16

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf KBlueLeaf/TIPO-100M:F16
# Run inference directly in the terminal:
./build/bin/llama cli -hf KBlueLeaf/TIPO-100M:F16

Use Docker

docker model run hf.co/KBlueLeaf/TIPO-100M:F16

Llama new
LM Studio
Jan

vLLM

How to use KBlueLeaf/TIPO-100M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KBlueLeaf/TIPO-100M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KBlueLeaf/TIPO-100M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/KBlueLeaf/TIPO-100M:F16

SGLang

How to use KBlueLeaf/TIPO-100M 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 "KBlueLeaf/TIPO-100M" \
    --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": "KBlueLeaf/TIPO-100M",
		"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 "KBlueLeaf/TIPO-100M" \
        --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": "KBlueLeaf/TIPO-100M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use KBlueLeaf/TIPO-100M with Ollama:
```
ollama run hf.co/KBlueLeaf/TIPO-100M:F16
```
Unsloth Desktop
Docker Model Runner
How to use KBlueLeaf/TIPO-100M with Docker Model Runner:
```
docker model run hf.co/KBlueLeaf/TIPO-100M:F16
```

Lemonade

How to use KBlueLeaf/TIPO-100M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull KBlueLeaf/TIPO-100M:F16

Run and chat with the model

lemonade run user.TIPO-100M-F16

List all available models

lemonade list

Atomic Chat

KBlueLeaf commited on Nov 29, 2024

Commit

371c622

verified ·

1 Parent(s): fd5085d

Update README.md

Browse files

Files changed (1) hide show

README.md +9 -10

README.md CHANGED Viewed

@@ -16,7 +16,7 @@ library_name: transformers
 ***This 100M model is still under development***
 100M LLaMA arch model trained for TIPO.<br>
-Tech Report: https://kblueleaf.net/document/TIPO-tech-report.pdf
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/630593e2fca1d8d92b81d2a1/fc9ovmARapQmgq9DZ7ApJ.png)
@@ -86,14 +86,13 @@ For research purpose, this model is released under Apache-2.0 License.
 ### Citation
 ```bibtex
-@misc{yeh2024tipo,
-  title = {TIPO: Text to Image with text presampling for Prompt Optimization},
-  author = {Yeh, Shih-Ying},
-  year = {2024},
-  month = {10},
-  day = {6},
-  note = {Technical report available at \url{https://kblueleaf.net/document/TIPO-tech-report.pdf}.
-          Model available at \url{https://huggingface.co/KBlueLeaf/TIPO-500M}.
-          Source code available at \url{https://github.com/KohakuBlueleaf/KGen}},
 }
 ```

 ***This 100M model is still under development***
 100M LLaMA arch model trained for TIPO.<br>
+Tech Report: https://arxiv.org/abs/2411.08127
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/630593e2fca1d8d92b81d2a1/fc9ovmARapQmgq9DZ7ApJ.png)
 ### Citation
 ```bibtex
+@misc{yeh2024tipotextimagetext,
+      title={TIPO: Text to Image with Text Presampling for Prompt Optimization},
+      author={Shih-Ying Yeh and Sang-Hyun Park and Giyeong Oh and Min Song and Youngjae Yu},
+      year={2024},
+      eprint={2411.08127},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2411.08127},
 }
 ```