Instructions to use spatialft/LFM2-350M-StepGame-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use spatialft/LFM2-350M-StepGame-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="spatialft/LFM2-350M-StepGame-GGUF",
	filename="LFM2-350M-StepGame-f16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use spatialft/LFM2-350M-StepGame-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf spatialft/LFM2-350M-StepGame-GGUF:F16
# Run inference directly in the terminal:
llama-cli -hf spatialft/LFM2-350M-StepGame-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf spatialft/LFM2-350M-StepGame-GGUF:F16
# Run inference directly in the terminal:
llama-cli -hf spatialft/LFM2-350M-StepGame-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 spatialft/LFM2-350M-StepGame-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf spatialft/LFM2-350M-StepGame-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 spatialft/LFM2-350M-StepGame-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf spatialft/LFM2-350M-StepGame-GGUF:F16

Use Docker

docker model run hf.co/spatialft/LFM2-350M-StepGame-GGUF:F16

LM Studio
Jan

vLLM

How to use spatialft/LFM2-350M-StepGame-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "spatialft/LFM2-350M-StepGame-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": "spatialft/LFM2-350M-StepGame-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/spatialft/LFM2-350M-StepGame-GGUF:F16

Ollama
How to use spatialft/LFM2-350M-StepGame-GGUF with Ollama:
```
ollama run hf.co/spatialft/LFM2-350M-StepGame-GGUF:F16
```

Unsloth Studio new

How to use spatialft/LFM2-350M-StepGame-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for spatialft/LFM2-350M-StepGame-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for spatialft/LFM2-350M-StepGame-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for spatialft/LFM2-350M-StepGame-GGUF to start chatting

Pi new

How to use spatialft/LFM2-350M-StepGame-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf spatialft/LFM2-350M-StepGame-GGUF:F16

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "spatialft/LFM2-350M-StepGame-GGUF:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use spatialft/LFM2-350M-StepGame-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf spatialft/LFM2-350M-StepGame-GGUF:F16

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default spatialft/LFM2-350M-StepGame-GGUF:F16

Run Hermes

hermes

Docker Model Runner
How to use spatialft/LFM2-350M-StepGame-GGUF with Docker Model Runner:
```
docker model run hf.co/spatialft/LFM2-350M-StepGame-GGUF:F16
```

Lemonade

How to use spatialft/LFM2-350M-StepGame-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull spatialft/LFM2-350M-StepGame-GGUF:F16

Run and chat with the model

lemonade run user.LFM2-350M-StepGame-GGUF-F16

List all available models

lemonade list

LFM2-350M-StepGame (GGUF)

Fine-tuned LiquidAI/LFM2-350M on the StepGame spatial reasoning benchmark. The model answers directional relationship questions (left, right, above, below, upper-left, upper-right, lower-left, lower-right) given a sequence of positional statements.

Results

Metric	Baseline	Fine-tuned	Delta
Overall	16.0%	74.4%	+58.4
1-hop	24.0%	94.0%	+70.0
2-hop	14.0%	90.0%	+76.0
3-hop	14.0%	76.0%	+62.0
4-hop	18.0%	54.0%	+36.0
5-hop	10.0%	58.0%	+48.0

Evaluated on 250 held-out examples (50 per hop level) from the StepGame validation split.

How to use

llama.cpp / llama-server

llama-server \
  --model LFM2-350M-StepGame-f16.gguf \
  --ctx-size 8192 \
  --host 0.0.0.0 --port 8080

llama-cpp-python

from llama_cpp import Llama

llm = Llama(model_path="LFM2-350M-StepGame-f16.gguf", n_ctx=8192)
output = llm.create_chat_completion(messages=[
    {"role": "system", "content": (
        "You are a spatial reasoning assistant. "
        "Given a sequence of positional relationships between objects, "
        "determine the spatial relationship between two specified objects. "
        "Answer with a single direction from: "
        "left, right, above, below, upper-left, upper-right, lower-left, lower-right."
    )},
    {"role": "user", "content": (
        "J and A are in a vertical line with A below J.\n\n"
        "What is the relation of the agent A to the agent J?"
    )},
])
print(output["choices"][0]["message"]["content"])
# => "below"

Training details

Parameter	Value
Base model	LiquidAI/LFM2-350M
Method	LoRA (PEFT)
Rank (r)	16
Alpha	32
Dropout	0.05
Target modules	q_proj, k_proj, v_proj, w1, w2, w3, in_proj, out_proj
Training examples	10,000 (2,000 per hop level, stratified)
Epochs	3
Learning rate	2e-4
Batch size	2 (x8 gradient accumulation)
Optimizer	paged_adamw_8bit
Quantization	QLoRA (NF4, double quant)
Final loss	0.2033
Training time	~77 min (Colab T4)

GGUF details

File	Quant	Size
`LFM2-350M-StepGame-f16.gguf`	F16	679 MB

Produced by merging the LoRA adapter into the base model, then converting with llama.cpp convert_hf_to_gguf.py.

Dataset

Training and evaluation data come from different splits of ZhengyanShi/StepGame:

Training: 10,000 examples from the train split (stratified, 2,000 per k-hop level)
Evaluation: 250 examples from the validation split (stratified, 50 per k-hop level)

Examples with the "overlap" label were filtered out. Only the 8 cardinal/intercardinal directions are used.

Prompt format

The model uses ChatML-style prompts (<|im_start|>/<|im_end|> tokens):

<|im_start|>system
You are a spatial reasoning assistant. Given a sequence of positional relationships between objects, determine the spatial relationship between two specified objects. Answer with a single direction from: left, right, above, below, upper-left, upper-right, lower-left, lower-right.<|im_end|>
<|im_start|>user
{story}

{question}<|im_end|>
<|im_start|>assistant

Source

Project repository: spatialft/spatialft.github.io

Built for AIPI 590.03 Intelligent Agents (Duke University).

Downloads last month: 15

GGUF

Model size

0.4B params

Architecture

lfm2

Hardware compatibility

16-bit

Model tree for spatialft/LFM2-350M-StepGame-GGUF

Base model

LiquidAI/LFM2-350M

Adapter

(18)

this model

Dataset used to train spatialft/LFM2-350M-StepGame-GGUF

Evaluation results

Overall Accuracy on StepGame (validation split)
validation set self-reported

74.400
1-hop Accuracy on StepGame (validation split)
validation set self-reported

94.000
2-hop Accuracy on StepGame (validation split)
validation set self-reported

90.000
3-hop Accuracy on StepGame (validation split)
validation set self-reported

76.000
4-hop Accuracy on StepGame (validation split)
validation set self-reported

54.000
5-hop Accuracy on StepGame (validation split)
validation set self-reported

58.000