Instructions to use tarpous/zero-125m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tarpous/zero-125m with Transformers:

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

pipe = pipeline("text-generation", model="tarpous/zero-125m")

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

tokenizer = AutoTokenizer.from_pretrained("tarpous/zero-125m")
model = AutoModelForCausalLM.from_pretrained("tarpous/zero-125m", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use tarpous/zero-125m 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 tarpous/zero-125m:F16
# Run inference directly in the terminal:
llama cli -hf tarpous/zero-125m:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf tarpous/zero-125m:F16
# Run inference directly in the terminal:
llama cli -hf tarpous/zero-125m:F16

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf tarpous/zero-125m:F16
# Run inference directly in the terminal:
./build/bin/llama cli -hf tarpous/zero-125m:F16

Use Docker

docker model run hf.co/tarpous/zero-125m:F16

Llama new
LM Studio
Jan

vLLM

How to use tarpous/zero-125m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tarpous/zero-125m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tarpous/zero-125m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tarpous/zero-125m:F16

SGLang

How to use tarpous/zero-125m 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 "tarpous/zero-125m" \
    --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": "tarpous/zero-125m",
		"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 "tarpous/zero-125m" \
        --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": "tarpous/zero-125m",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use tarpous/zero-125m with Ollama:
```
ollama run hf.co/tarpous/zero-125m:F16
```
Unsloth Desktop
Docker Model Runner
How to use tarpous/zero-125m with Docker Model Runner:
```
docker model run hf.co/tarpous/zero-125m:F16
```

Lemonade

How to use tarpous/zero-125m with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tarpous/zero-125m:F16

Run and chat with the model

lemonade run user.zero-125m-F16

List all available models

lemonade list

Atomic Chat

zero-125m / evaluation /zero-125m-validation.jsonl

tarpous

Add zero-125m base model and GGUF artifacts

504625d verified 3 months ago

Raw History Blame Contribute Delete

1.46 kB

{"batch_count":32,"cursor_after":{"build_manifest_sha256":"27a6fd81c772c0384c1d0b43a102df7142de081869234ec5607f7370729c2c1b","epoch":0,"format":"lm-from-zero-batch-cursor","format_version":1,"next_local_window":32,"rank":0,"seed":1337,"sequence_length":2048,"sequences_consumed":32,"shuffle":false,"split":"validation","tokenizer_hash":"e16b820d0d8c7ad8e89904714ef6d89eb573a4e651dd337bde56fd8f418476dd","tokens_consumed":65536,"world_size":1},"cursor_before":{"build_manifest_sha256":"27a6fd81c772c0384c1d0b43a102df7142de081869234ec5607f7370729c2c1b","epoch":0,"format":"lm-from-zero-batch-cursor","format_version":1,"next_local_window":0,"rank":0,"seed":1337,"sequence_length":2048,"sequences_consumed":0,"shuffle":false,"split":"validation","tokenizer_hash":"e16b820d0d8c7ad8e89904714ef6d89eb573a4e651dd337bde56fd8f418476dd","tokens_consumed":0,"world_size":1},"elapsed_seconds":0.7582138660000055,"evaluated_at_utc":"2026-08-09T18:30:41.479327Z","format":"lm-from-zero-causal-evaluation","format_version":1,"mean_loss":3.3249846026301384,"model_config_sha256":"c886f7c13c6c027276c3f2ddcede72425fac511e2d1b2551b5d7eadd121f2bf4","perplexity":27.798570614501042,"predicted_token_count":65504,"predicted_tokens_per_second":86392.51131817144,"sequence_count":32,"shard_manifest_sha256":"27a6fd81c772c0384c1d0b43a102df7142de081869234ec5607f7370729c2c1b","split":"validation","tokenizer_sha256":"e16b820d0d8c7ad8e89904714ef6d89eb573a4e651dd337bde56fd8f418476dd"}