Instructions to use RichardErkhov/Locutusque_-_TinyMistral-248M-8bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RichardErkhov/Locutusque_-_TinyMistral-248M-8bits with Transformers:

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

pipe = pipeline("text-generation", model="RichardErkhov/Locutusque_-_TinyMistral-248M-8bits")

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

tokenizer = AutoTokenizer.from_pretrained("RichardErkhov/Locutusque_-_TinyMistral-248M-8bits")
model = AutoModelForCausalLM.from_pretrained("RichardErkhov/Locutusque_-_TinyMistral-248M-8bits", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RichardErkhov/Locutusque_-_TinyMistral-248M-8bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RichardErkhov/Locutusque_-_TinyMistral-248M-8bits

SGLang

How to use RichardErkhov/Locutusque_-_TinyMistral-248M-8bits 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 "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits" \
    --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": "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits",
		"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 "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits" \
        --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": "RichardErkhov/Locutusque_-_TinyMistral-248M-8bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RichardErkhov/Locutusque_-_TinyMistral-248M-8bits with Docker Model Runner:
```
docker model run hf.co/RichardErkhov/Locutusque_-_TinyMistral-248M-8bits
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Quantization made by Richard Erkhov.

Github

Discord

Request more models

TinyMistral-248M - bnb 8bits

Model creator: https://huggingface.co/Locutusque/
Original model: https://huggingface.co/Locutusque/TinyMistral-248M/

Original model description:

license: apache-2.0 datasets: - Skylion007/openwebtext - JeanKaddour/minipile language: - en pipeline_tag: text-generation inference: parameters: do_sample: True temperature: 0.5 top_p: 0.5 top_k: 50 max_new_tokens: 250 repetition_penalty: 1.176

A pre-trained language model, based on the Mistral 7B model, has been scaled down to approximately 248 million parameters. This model has been trained on 7,488,000 examples. This model isn't intended for direct use but for fine-tuning on a downstream task. This model should have a context length of around 32,768 tokens. Safe serialization has been removed due to issues saving model weights.

During evaluation on InstructMix, this model achieved an average perplexity score of 6.3. More epochs are planned for this model on different datasets.

Open LLM Leaderboard Evaluation Results (outdated)

Detailed results can be found here

Metric	Value
Avg.	24.18
ARC (25-shot)	20.82
HellaSwag (10-shot)	26.98
MMLU (5-shot)	23.11
TruthfulQA (0-shot)	46.89
Winogrande (5-shot)	50.75
GSM8K (5-shot)	0.0
DROP (3-shot)	0.74

The purpose of this model is to prove that trillion-scale datasets are not needed to pretrain a language model. As a result of needing small datasets, this model was pretrained on a single GPU (Titan V).

Downloads last month: 7

Safetensors

Model size

0.2B params

Tensor type

F32

F16