Instructions to use gggff123/lfmCodeEnhanced with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use gggff123/lfmCodeEnhanced 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 gggff123/lfmCodeEnhanced:Q4_K_M # Run inference directly in the terminal: llama cli -hf gggff123/lfmCodeEnhanced:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf gggff123/lfmCodeEnhanced:Q4_K_M # Run inference directly in the terminal: llama cli -hf gggff123/lfmCodeEnhanced: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 gggff123/lfmCodeEnhanced:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf gggff123/lfmCodeEnhanced:Q4_K_M
Use Docker
docker model run hf.co/gggff123/lfmCodeEnhanced:Q4_K_M
- Llama new
- LM Studio
- Jan
- Ollama
How to use gggff123/lfmCodeEnhanced with Ollama:
ollama run hf.co/gggff123/lfmCodeEnhanced:Q4_K_M
- Unsloth Desktop
- Pi
How to use gggff123/lfmCodeEnhanced with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf gggff123/lfmCodeEnhanced: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": "gggff123/lfmCodeEnhanced:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use gggff123/lfmCodeEnhanced with Docker Model Runner:
docker model run hf.co/gggff123/lfmCodeEnhanced:Q4_K_M
- Lemonade
How to use gggff123/lfmCodeEnhanced with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull gggff123/lfmCodeEnhanced:Q4_K_M
Run and chat with the model
lemonade run user.lfmCodeEnhanced-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use gggff123/lfmCodeEnhanced with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf gggff123/lfmCodeEnhanced: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 gggff123/lfmCodeEnhanced:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use gggff123/lfmCodeEnhanced with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf gggff123/lfmCodeEnhanced: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 "gggff123/lfmCodeEnhanced: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"
File size: 2,521 Bytes
e647259 5a0aa6b e647259 5a0aa6b 8d0b89e 5a0aa6b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 | ---
license: mit
language:
- en
base_model:
- LiquidAI/LFM2.5-230M
---
## LFM 2.5 230m Code-Enhanced
Lfm 2.5 230m code enhanced is a code-ish version of the previous lfm 2.5 230m (base) , this model however attempts to make the old model better at coding
this uses the latest weight space programming paper along with deepseek v4.1 flash to manually change the model weights.
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>LFM CodeEnhanced AI</title>
<style>
* { box-sizing: border-box; }
body {
margin: 0;
min-height: 100vh;
display: grid;
place-items: center;
background: #080b14;
font-family: Inter, Arial, sans-serif;
}
.logo {
display: flex;
align-items: center;
gap: 18px;
}
svg { width: 100px; height: 100px; }
.brand-name {
font-size: 30px;
font-weight: 850;
letter-spacing: -1.5px;
color: black;
}
.brand-name span {
background: linear-gradient(100deg,#00e5ff,#9361ff);
-webkit-background-clip: text;
background-clip: text;
color: transparent;
}
.tagline {
margin-top: 7px;
color: #8996b5;
font-size: 10px;
font-weight: 700;
letter-spacing: 4px;
}
</style>
</head>
<body>
<div class="logo">
<svg viewBox="0 0 100 100" fill="none">
<defs>
<linearGradient id="g" x1="10" y1="10" x2="90" y2="90">
<stop stop-color="#00e5ff"/>
<stop offset="1" stop-color="#9361ff"/>
</linearGradient>
</defs>
<path d="M50 4 91 27V73L50 96 9 73V27L50 4Z"
fill="#111a30" stroke="url(#g)" stroke-width="2.5"/>
<path d="m35 33-15 17 15 17M65 33l15 17-15 17"
stroke="url(#g)" stroke-width="5"
stroke-linecap="round" stroke-linejoin="round"/>
<path d="m50 23 6 21 21 6-21 6-6 21-6-21-21-6 21-6 6-21Z"
fill="url(#g)"/>
<circle cx="50" cy="50" r="5" fill="white"/>
</svg>
<div>
<div class="brand-name">LFM <span>CodeEnhanced</span></div>
<div class="tagline">AI · CODE · CREATE</div>
</div>
</div>
</body>
</html>
Currently has 3 quantizations
--
| Quantization | Quality |
|----------------|--------|
| BF16 | Highest quality <span style="background-color: #12CA12; color: black; padding: 2px 8px; border-radius: 10px; font-size: 12px; font-weight: bold;">Recommended</span>
| Q4_KM | Mid Quality |
| Q8 | Also good |
---
This model was generated by deepseek V4.1 FLASH by changing model weights.
All credits to <b>liquid-ai, deepseek & weight space programming research paper</b>
--- |