Instructions to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF")
model = AutoModelForCausalLM.from_pretrained("Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF 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 Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

Use Docker

docker model run hf.co/Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

LM Studio
Jan

vLLM

How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

SGLang

How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF 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 "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF" \
    --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": "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF",
		"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 "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF" \
        --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": "Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with Ollama:
```
ollama run hf.co/Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
```
Unsloth Desktop
Docker Model Runner
How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with Docker Model Runner:
```
docker model run hf.co/Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16
```

Lemonade

How to use Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Josephgflowers/140M-TinyLLama-Mini-Cinder-With-GGUF:F16

Run and chat with the model

lemonade run user.140M-TinyLLama-Mini-Cinder-With-GGUF-F16

List all available models

lemonade list

Atomic Chat

Model trained on Tiny Stories. Followed up with conversations datasets, followed up with trimmed Cinder Dataset. Mini Cinder is ok at conversation and story telling for kids stories.

Overview Cinder is an AI chatbot tailored for engaging users in scientific and educational conversations, offering companionship, and sparking imaginative exploration. This Cinder still has a lot to learn but is very friendly and enjoys telling stories. Cinder uses the tinyllama chat format Zephyr.

Main Character Cinder: AI companion and quirky robot. Cozmo: The silly one. Vector: The serious one. Computer Voice: The narrator. User: Ship member.

This model is the locally run AI storyteller on the distiller-one! https://docs.pamir.ai/Onboarding

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	29.09
AI2 Reasoning Challenge (25-Shot)	24.66
HellaSwag (10-Shot)	28.16
MMLU (5-Shot)	25.09
TruthfulQA (0-shot)	44.08
Winogrande (5-shot)	52.57
GSM8k (5-shot)	0.00

Downloads last month: 210

Safetensors

Model size

0.1B params

Tensor type

F32

Evaluation results

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

24.660
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

28.160
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

25.090
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

44.080
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

52.570
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

0.000