Instructions to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-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 Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

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 Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Nullmega/LFM2.5-8B-A1B-pi-coding-security-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": "Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Ollama
How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with Ollama:
```
ollama run hf.co/Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M
```
Unsloth Desktop

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with Docker Model Runner:
```
docker model run hf.co/Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M
```

Lemonade

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-8B-A1B-pi-coding-security-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

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 Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

LFM2.5-8B-A1B pi coding and defensive-security experiment

User-trained derivative of LiquidAI/LFM2.5-8B-A1B, pinned base revision dd83f5cb02aa2938b66bba9a9998d38d0517c1f3. Experimental fine-tune: broader coding, security and retention gains have not been established.

Training

One complete pass: 126 optimizer updates, 656,800 input tokens, 210,697 supervised assistant tokens, 504 examples. BF16 frozen base and FP32 LoRA, rank16/alpha32/dropout0.05 on66 attention, convolution-projection and shallow dense-FFN modules. Experts, routers, tied embeddings/head and norms were frozen. No sequence packing. Sequence limit2048. Actual GPU: NVIDIA L4; A100 was unavailable.

The earlier2M-token plan was an upper budget. Licensing and execution checks reduced the accepted corpus; data was not repeated to fill that budget. Sources and revisions are in training_data_attribution.json. Tool demonstrations were driven by a deterministic fixture oracle and executed by real pi1.5.0; they are not target-model success claims.

Compact paired evaluation

Original: 0/8 protected pi tasks, 2/4 instruction checks. Trained BF16 adapter: 2/8 protected pi tasks, 4/4 instruction checks. These small clustered diagnostic cases do not establish a general improvement percentage. The broader proposed96-task final suite was not run. The diagnostic bridge permits file tools only on app.py and a small fixed command list. An out-of-scope call is rejected at the bridge and can terminate the case rather than returning a normal pi tool error; this limits conclusions about recovery and real repositories. Instruction probes use a 128-token output budget and include a task-directory marker in the system context. See evaluation_summary.json for aggregate results.

The token vocabulary is unchanged. The versioned corrected-v2 chat template omits empty tool-call blocks, escapes Pythonic argument values, normalizes tool wrappers and defaults to an explicit empty reasoning block. The identical template was used to prepare training data, serve the matched comparison and export GGUF. This is an experimental runtime change, so the paired scores above apply to that shared runtime rather than the unmodified template. LFM2.5 uses native Pythonic tool calls; pi requires a tested tool-format conversion and parser. GGUF load/generation checks do not establish an entire pi serving round trip.

License and attribution

See LICENSE, NOTICE and training_data_attribution.json. The base LFM Open License v1.0 includes commercial-use conditions and is not Apache-2.0. Modifications are the trained adapter, its merge, the corrected-v2 chat template and the named quantizations. Upstream source notices are retained. Raw private sessions, optimizer state and protected task content are not included.

Files

All four files derive from the same trained merge. Q2_K is the most lossy and is provided as requested; per-file checks are reported rather than assumed equivalent quality. The exact two generated outputs for each file are included in generation-samples-QUANT.json. A failed case-sensitive READY check can reflect a capitalization difference; the recorded result remains a failure.

Quant	File	Size	READY check	JSON check
Q8_0	LFM2.5-8B-A1B-pi-trained-Q8_0.gguf	8.39 GiB	False	True
Q6_K	LFM2.5-8B-A1B-pi-trained-Q6_K.gguf	6.48 GiB	False	True
Q4_K_M	LFM2.5-8B-A1B-pi-trained-Q4_K_M.gguf	4.80 GiB	False	True
Q2_K	LFM2.5-8B-A1B-pi-trained-Q2_K.gguf	2.97 GiB	False	True

Conversion used pinned llama.cpp 5851a8e505f4d5fbb183f2847698eb6155d176b7. Native architecture: lfm2moe. Load checks used context2048; larger fine-tuned contexts were not validated.

Downloads last month: 215

GGUF

Model size

8B params

Architecture

lfm2moe

Hardware compatibility

2-bit

4-bit

6-bit

8-bit

Model tree for Nullmega/LFM2.5-8B-A1B-pi-coding-security-GGUF

Base model

LiquidAI/LFM2.5-8B-A1B-Base

Finetuned

LiquidAI/LFM2.5-8B-A1B

Quantized

(106)

this model