Instructions to use vicgalle/Miqu-6B-truthy with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use vicgalle/Miqu-6B-truthy with Transformers:

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

pipe = pipeline("text-generation", model="vicgalle/Miqu-6B-truthy")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("vicgalle/Miqu-6B-truthy")
model = AutoModelForMultimodalLM.from_pretrained("vicgalle/Miqu-6B-truthy")
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 vicgalle/Miqu-6B-truthy with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "vicgalle/Miqu-6B-truthy"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "vicgalle/Miqu-6B-truthy",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/vicgalle/Miqu-6B-truthy

SGLang

How to use vicgalle/Miqu-6B-truthy 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 "vicgalle/Miqu-6B-truthy" \
    --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": "vicgalle/Miqu-6B-truthy",
		"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 "vicgalle/Miqu-6B-truthy" \
        --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": "vicgalle/Miqu-6B-truthy",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use vicgalle/Miqu-6B-truthy with Docker Model Runner:
```
docker model run hf.co/vicgalle/Miqu-6B-truthy
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Miqu-6B-truthy

A truthfully Miqu of 6B parameters, as an experiment.

"results": {
    "truthfulqa_mc": {
      "mc1": 0.2521419828641371,
      "mc1_stderr": 0.01520152224629995,
      "mc2": 0.5051887026752994,
      "mc2_stderr": 0.016738600540275827
    }
  },

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	30.28
AI2 Reasoning Challenge (25-Shot)	27.65
HellaSwag (10-Shot)	26.71
MMLU (5-Shot)	27.04
TruthfulQA (0-shot)	50.63
Winogrande (5-shot)	49.64
GSM8k (5-shot)	0.00

Downloads last month: 89

Safetensors

Model size

6B params

Tensor type

F16

Model tree for vicgalle/Miqu-6B-truthy

Quantizations

2 models

Dataset used to train vicgalle/Miqu-6B-truthy

Evaluation results

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

27.650
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

26.710
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

27.040
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

50.630
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

49.640
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

0.000