Instructions to use migtissera/Tess-XS-v1-3-yarn-128K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use migtissera/Tess-XS-v1-3-yarn-128K with Transformers:

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

pipe = pipeline("text-generation", model="migtissera/Tess-XS-v1-3-yarn-128K", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("migtissera/Tess-XS-v1-3-yarn-128K", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("migtissera/Tess-XS-v1-3-yarn-128K", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use migtissera/Tess-XS-v1-3-yarn-128K with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "migtissera/Tess-XS-v1-3-yarn-128K"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "migtissera/Tess-XS-v1-3-yarn-128K",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/migtissera/Tess-XS-v1-3-yarn-128K

SGLang

How to use migtissera/Tess-XS-v1-3-yarn-128K 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 "migtissera/Tess-XS-v1-3-yarn-128K" \
    --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": "migtissera/Tess-XS-v1-3-yarn-128K",
		"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 "migtissera/Tess-XS-v1-3-yarn-128K" \
        --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": "migtissera/Tess-XS-v1-3-yarn-128K",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use migtissera/Tess-XS-v1-3-yarn-128K with Docker Model Runner:
```
docker model run hf.co/migtissera/Tess-XS-v1-3-yarn-128K
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Note:

This version is the stable release. The issues that were present in versions 1.0, 1.1 and 1.2 all have been rectified. Thank you for your patience while R&D was conducted. Enjoy!

This model have been tested on context length up to 16K. Model produced slight repetition around 16K context length. I recommend testing the model to your usecase and limiting the context length.

Here's my learnings going from Tess-v1.0 to Tess-v1.3: https://migel.substack.com/p/learnings-from-training-tess

Tess

Tess, short for Tesoro (Treasure in Italian), is a general purpose Large Language Model series. Tess-XS-v1.3 was trained on the Nous Research Mistral-7B-yarn-128K base.

Prompt Format:

SYSTEM: <ANY SYSTEM CONTEXT>
USER: 
ASSISTANT:

Downloads last month: 785

Model tree for migtissera/Tess-XS-v1-3-yarn-128K

Finetunes

6 models

Merges

1 model

Quantizations

3 models