Instructions to use t3nchu/mikrotik-expert 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 t3nchu/mikrotik-expert 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 t3nchu/mikrotik-expert:Q4_K_M # Run inference directly in the terminal: llama cli -hf t3nchu/mikrotik-expert:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf t3nchu/mikrotik-expert:Q4_K_M # Run inference directly in the terminal: llama cli -hf t3nchu/mikrotik-expert: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 t3nchu/mikrotik-expert:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf t3nchu/mikrotik-expert: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 t3nchu/mikrotik-expert:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf t3nchu/mikrotik-expert:Q4_K_M
Use Docker
docker model run hf.co/t3nchu/mikrotik-expert:Q4_K_M
- LM Studio
- Jan
- Ollama
How to use t3nchu/mikrotik-expert with Ollama:
ollama run hf.co/t3nchu/mikrotik-expert:Q4_K_M
- Unsloth Desktop
- Docker Model Runner
How to use t3nchu/mikrotik-expert with Docker Model Runner:
docker model run hf.co/t3nchu/mikrotik-expert:Q4_K_M
- Lemonade
How to use t3nchu/mikrotik-expert with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull t3nchu/mikrotik-expert:Q4_K_M
Run and chat with the model
lemonade run user.mikrotik-expert-Q4_K_M
List all available models
lemonade list
- Atomic Chat
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
MikroTik RouterOS Expert - Ollama Model
A fine-tuned DeepSeek-R1-Distill-Llama-8B model specialized in MikroTik RouterOS 7.x configuration, troubleshooting, and command reference.
Requirements
- Ollama installed
- RAM/VRAM per the tier you choose (see below)
Available Tiers
| Tier | File | Size | Min RAM | Best for |
|---|---|---|---|---|
| Premium | mikrotik-expert-q8_0.gguf |
8.0 GB | 12 GB | Best quality, 16GB+ VRAM GPUs |
| Standard | mikrotik-expert-q4_k_m.gguf |
4.6 GB | 6 GB | Recommended default, any modern GPU or CPU |
| Lite | mikrotik-expert-q3_k_m.gguf |
3.8 GB | 4 GB | Low-end laptops, Raspberry Pi, minimal setups |
Not sure which to pick? Go with Standard (Q4_K_M) — it runs on almost anything and the quality loss vs Q8_0 is minimal for RouterOS tasks.
Install
- Download your chosen
.gguffile and its matchingModelfile.<tier>into the same folder - Rename the Modelfile:
# Example for Standard tier:
cp Modelfile.q4_k_m Modelfile
- Import into Ollama:
ollama create mikrotik-expert -f Modelfile
- Start chatting:
ollama run mikrotik-expert
Model Details
| Property | Value |
|---|---|
| Base model | DeepSeek-R1-Distill-Llama-8B |
| Fine-tune method | LoRA (rank 64, alpha 128) |
| Training data | ~9,200 MikroTik RouterOS instruction pairs |
| Data sources | MikroTik wiki, forums, GitHub repos, HuggingFace datasets, synthetic Q&A |
| Context length | 2048 tokens |
What it knows
- RouterOS 7.x CLI commands and syntax
- Firewall rules, NAT, mangle
- OSPF, BGP, MPLS routing
- Wireless (WiFi, CAPsMAN)
- VPN (WireGuard, IPsec, L2TP, SSTP)
- QoS, queues, traffic shaping
- Scripting and scheduling
- Bridge, VLAN, bonding
- DHCP, DNS, certificates
- Troubleshooting common issues
Example
>>> How do I set up a basic firewall on RouterOS 7?
<think>
The user wants a basic firewall setup for RouterOS 7...
</think>
/ip firewall filter
add chain=input connection-state=established,related action=accept
add chain=input connection-state=invalid action=drop
add chain=input protocol=icmp action=accept
add chain=input src-address=192.168.88.0/24 action=accept
add chain=input action=drop
License
Private - Raiden Tech FZ LLC.
- Downloads last month
- 2
Hardware compatibility
Log In to add your hardware
3-bit
4-bit
8-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support