Instructions to use migtissera/Tess-72B-v1.5b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use migtissera/Tess-72B-v1.5b with Transformers:

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

pipe = pipeline("text-generation", model="migtissera/Tess-72B-v1.5b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("migtissera/Tess-72B-v1.5b")
model = AutoModelForCausalLM.from_pretrained("migtissera/Tess-72B-v1.5b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use migtissera/Tess-72B-v1.5b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "migtissera/Tess-72B-v1.5b"
# 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-72B-v1.5b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/migtissera/Tess-72B-v1.5b

SGLang

How to use migtissera/Tess-72B-v1.5b 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-72B-v1.5b" \
    --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-72B-v1.5b",
		"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-72B-v1.5b" \
        --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-72B-v1.5b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use migtissera/Tess-72B-v1.5b with Docker Model Runner:
```
docker model run hf.co/migtissera/Tess-72B-v1.5b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Tess, short for Tesoro (Treasure in Italian), is a general purpose Large Language Model series. Tess-72B-v1.5b was trained on the Qwen-72B base.

Prompt Format:

SYSTEM: <ANY SYSTEM CONTEXT>
USER: 
ASSISTANT:

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	77.30
AI2 Reasoning Challenge (25-Shot)	71.25
HellaSwag (10-Shot)	85.53
MMLU (5-Shot)	76.63
TruthfulQA (0-shot)	71.99
Winogrande (5-shot)	81.45
GSM8k (5-shot)	76.95

Downloads last month: 73

Safetensors

Model size

72B params

Tensor type

F16

Model tree for migtissera/Tess-72B-v1.5b

Merges

1 model

Quantizations

2 models

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

71.250
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

85.530
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

76.630
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

71.990
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

81.450
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

76.950