Instructions to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF 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 SONGSSONG123/gemma4-26b-cafe-64e-text-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 SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf SONGSSONG123/gemma4-26b-cafe-64e-text-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 SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf SONGSSONG123/gemma4-26b-cafe-64e-text-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 SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama cli -hf SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
Use Docker
docker model run hf.co/SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
- Llama new
- LM Studio
- Jan
- vLLM
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "SONGSSONG123/gemma4-26b-cafe-64e-text-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": "SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
- Ollama
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with Ollama:
ollama run hf.co/SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
- Unsloth Desktop
- Pi
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf SONGSSONG123/gemma4-26b-cafe-64e-text-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:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with Docker Model Runner:
docker model run hf.co/SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
- Lemonade
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.gemma4-26b-cafe-64e-text-GGUF-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-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 SONGSSONG123/gemma4-26b-cafe-64e-text-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://127.0.0.1:9931/v1 hermes config set model.default SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf SONGSSONG123/gemma4-26b-cafe-64e-text-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://127.0.0.1:9931/v1 \ --custom-model-id "SONGSSONG123/gemma4-26b-cafe-64e-text-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"
gemma4-26b-cafe-64e-text (GGUF, Q4_K_M)
An experiment: half of the experts of Gemma 4 26B (MoE, 128 experts Γ 30 layers) were removed using Korean cafΓ©-order calibration data, so the model fits on a laptop integrated GPU while keeping cafΓ© ordering skills.
This is a domain-specific experimental model, not a general-purpose model. General knowledge is significantly reduced.
What was changed
- Base: google/gemma-4-26B-A4B-it (Apache 2.0)
- Structured expert pruning: 128 β 64 experts per layer (1,920 of 3,840 experts removed). Remaining weights are unchanged; expert tensors, router rows, and per-expert scales were sliced. Top-k routing stays 8.
- Expert selection: per-expert contribution sum Ξ£ gΒ·βf(x)β over 600 Korean cafΓ©-order prompts (system prompt + tool definitions + customer utterance), fixed prompt and request tokens weighted 50:50, plus protection for experts handling β₯30% of a layer's tokens. Based on REAP (Lasby et al., 2025, arXiv:2510.13999), modified to account for routing frequency.
- Quantized to Q4_K_M (per-tensor types matched to the original Ollama build), no imatrix.
- Text only: vision components removed. Language weights are identical to the full 64e build used in the evaluation below.
Results (single run, thinking off, temperature 0)
| Model | CafΓ© test (120, AI-written) | Live test (20, human-written) | General knowledge (KMMLU 200 + MMLU 100) |
|---|---|---|---|
| Gemma 4 26B original | 94.2% | 75% | 72.7% |
| 64e (this model) | 93.3% | 70% | 47.3% |
| Gemma 4 12B | 90.8% | 45% | 64.3% |
Same laptop integrated GPU (Intel Arc 140T): the full 64e build (with Ollama's MTP draft model) ran at a median 2.55 s per order (Gemma 4 12B, also with its draft model: 3.33 s). This text-only GGUF does not include the draft model: median 6.46 s per order. The original 26B does not fit on this iGPU.
Limitations
- Live test has only 20 items; no comparison is statistically significant (p β₯ 0.05).
- CafΓ© test, calibration data, and prompts were generated with an AI coding tool; only the live test was human-written.
- Tuned for a specific cafΓ© menu/tool schema. Expect weak performance outside cafΓ© ordering.
- Image input is not supported and was never evaluated.
Usage
ollama run hf.co/SONGSSONG123/gemma4-26b-cafe-64e-text-GGUF:Q4_K_M
Or download the GGUF and ollama create cafe-64e -f Modelfile.
Links
- Code and results: https://github.com/gustjrvkdnj1234-png/cafe-robot-brain
- Blog post (Korean): https://blog.cslee.co.kr/a-graduate-ai-applied-for-a-cafe-part-time-job
License
Apache 2.0. This is a modified version of Google's Gemma 4; see LICENSE.
- Downloads last month
- 502
4-bit