Instructions to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

Use Docker

docker model run hf.co/RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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": "RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

Ollama
How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with Ollama:
```
ollama run hf.co/RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
```
Unsloth Desktop

How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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": "RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with Docker Model Runner:
```
docker model run hf.co/RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M
```

Lemonade

How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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 "RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA-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"

LFM2.5-1.2B-Thinking-LeetCode-QLoRA — GGUF

Quantized GGUF builds of the LoRA adapter RenShiPDev/LFM2.5-1.2B-Thinking-LeetCode-QLoRA, merged into LiquidAI/LFM2.5-1.2B-Thinking.

QLoRA fine-tune (500 steps on greengerong/leetcode), converted for CPU / llama.cpp inference. Research checkpoint, not a production model.

All quants were produced from the BF16 merge with llama-quantize (no requantization).

Quality vs BF16

Measured on WikiText-2 (test split, ~246k tokens) with llama-perplexity --kl-divergence, teacher-forced. BF16 is the reference: lower PPL is better, and KLD mean is the mean KL divergence between this quant's token distribution and BF16 (0 = identical).

Quant	Size	bpw	PPL ↓	KLD mean ↓
BF16	2.2 GB	16.00	22.9431	0.00000
F16	2.2 GB	16.00	22.9700	0.00017
Q8_0	1.2 GB	8.51	22.9140	0.00129
Q6_K	918.2 MB	6.57	23.1350	0.00524
Q5_K_M	804.3 MB	5.76	23.6043	0.01776
Q4_K_M	697.0 MB	4.99	23.3247	0.04798
IQ4_XS	632.6 MB	4.53	23.7408	0.06506
Q3_K_M	572.5 MB	4.10	25.2301	0.15033
Q2_K_L	492.0 MB	3.52	34.3705	0.52051
IQ3_XXS	468.2 MB	3.35	26.2662	0.30569
IQ2_M	414.0 MB	2.96	32.2434	0.50539
IQ2_XXS	348.2 MB	2.49	66.4492	1.23602

Recommendation: Q4_K_M — best size/quality trade-off (~697 MB, +0.38 PPL, KLD 0.048). Q6_K/Q8_0 are near-lossless. Below Q3_K_M degradation becomes large.

How it was measured

# 1. save reference logits (once)
llama-perplexity -m LFM2.5-1.2B-Thinking.BF16.gguf -f wiki.test.raw -c 512 -ngl 99 \
  --save-all-logits base_logits.bin

# 2. PPL + KL divergence per quant
llama-perplexity -m <quant>.gguf -f wiki.test.raw -c 512 -ngl 99 \
  --kl-divergence --kl-divergence-base base_logits.bin

Usage

# CLI
llama-cli -m LFM2.5-1.2B-Thinking.Q4_K_M.gguf -p "Explain binary search."

# Server (OpenAI-compatible)
llama-server -m LFM2.5-1.2B-Thinking.Q4_K_M.gguf --port 8080