Sample Page Title

Picture by Writer | Ideogram

Working a number of massive language fashions might be helpful, whether or not for evaluating mannequin outputs, establishing a fallback in case one fails, or customizing habits (like utilizing one mannequin for coding and one other for technical writing). That is how we regularly use LLMs in apply. There are apps like poe.com that supply this type of setup. It’s a single platform the place you’ll be able to run a number of LLMs. However what if you wish to do all of it regionally, save on API prices, and maintain your information personal?

Nicely, that’s the place the true downside exhibits up. Setting this up often means juggling completely different ports, operating separate processes, and switching between them manually. Not ideally suited.

That’s precisely the ache Llama-Swap solves. It’s an open-source proxy server that’s tremendous light-weight (only a single binary), and it permits you to swap between a number of native LLMs simply. In easy phrases, it listens for OpenAI-style API calls in your machine and robotically begins or stops the suitable mannequin server primarily based on the mannequin you request. Let’s break down the way it works and stroll by means of a step-by-step setup to get it operating in your native machine.

# How Llama-Swap Works

 
Conceptually, Llama-Swap sits in entrance of your LLM servers as a wise router. When an API request arrives (e.g., a POST /v1/chat/completions name), it seems on the "mannequin" discipline within the JSON payload. It then hundreds the suitable server course of for that mannequin, shutting down every other mannequin if wanted. For instance, should you first request mannequin "A" after which request mannequin "B", Llama-Swap will robotically cease the server for “A” and begin the server for “B” so that every request is served by the right mannequin. This dynamic swapping occurs transparently, so shoppers see the anticipated response with out worrying in regards to the underlying processes.

By default, Llama-Swap permits just one mannequin to run at a time (it unloads others when switching). Nevertheless, its Teams function permits you to change this habits. A gaggle can record a number of fashions and management their swap habits. For instance, setting swap: false in a gaggle means all group members can run collectively with out unloading. In apply, you would possibly use one group for heavyweight fashions (just one energetic at a time) and one other “parallel” group for small fashions you need operating concurrently. This provides you full management over useful resource utilization and concurrency on a single server.

# Conditions

 
Earlier than getting began, guarantee your system has the next:

• Python 3 (>=3.8): Wanted for fundamental scripting and tooling.
• Homebrew (on macOS): Makes putting in LLM runtimes straightforward. For instance, you’ll be able to set up the llama.cpp server with:

This gives the llama-server binary for internet hosting fashions regionally.

• llama.cpp (llama-server): The OpenAI-compatible server binary (put in through Homebrew above, or constructed from supply) that really runs the LLM mannequin.
• Hugging Face CLI: For downloading fashions on to your native machine with out logging into the positioning or manually navigating mannequin pages. Set up it utilizing:

pip set up -U "huggingface_hub[cli]"

• {Hardware}: Any fashionable CPU will work. For quicker inference, a GPU is beneficial. (On Apple Silicon Macs, you’ll be able to run on the CPU or strive PyTorch’s MPS backend for supported fashions. On Linux/Home windows with NVIDIA GPUs, you should use Docker/CUDA containers for acceleration.)
• Docker (Non-compulsory): To run the pre-built Docker photos. Nevertheless, I selected to not use this for this information as a result of these photos are designed primarily for x86 (Intel/AMD) programs and don’t work reliably on Apple Silicon (M1/M2) Macs. As an alternative, I used the bare-metal set up technique, which works immediately on macOS with none container overhead.

In abstract, you’ll want a Python setting and a neighborhood LLM server (just like the `llama.cpp` server). We’ll use these to host two instance fashions on one machine.

# Step-by-Step Directions

// 1. Putting in Llama-Swap

Obtain the newest Llama-Swap launch on your OS from the GitHub releases web page. For instance, I may see v126 as the newest launch. Run the next instructions:

# Step 1: Obtain the right file
curl -L -o llama-swap.tar.gz 
  https://github.com/mostlygeek/llama-swap/releases/obtain/v126/llama-swap_126_darwin_arm64.tar.gz

Output:
  % Whole    % Acquired % Xferd  Common Velocity   Time    Time     Time  Present
                                 Dload  Add   Whole   Spent    Left  Velocity
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
100 3445k  100 3445k    0     0  1283k      0  0:00:02  0:00:02 --:--:-- 5417k

Now, extract the file, make it executable, and check it by checking the model:

# Step 2: Extract it
tar -xzf llama-swap.tar.gz

# Step 3: Make it executable
chmod +x llama-swap

# Step 4: Take a look at it
./llama-swap --version

Output:
model: 126 (591a9cdf4d3314fe4b3906e939a17e76402e1655), constructed at 2025-06-16T23:53:50Z

// 2. Downloading and Getting ready Two or Extra LLMs

Select two instance fashions to run. We’ll use Qwen2.5-0.5B and SmolLM2-135M (small fashions) from Hugging Face. You want the mannequin information (in GGUF or related format) in your machine. For instance, utilizing the Hugging Face CLI:

mkdir -p ~/llm-models

huggingface-cli obtain bartowski/SmolLM2-135M-Instruct-GGUF 
  --include "SmolLM2-135M-Instruct-Q4_K_M.gguf" --local-dir ~/llm-models

huggingface-cli obtain bartowski/Qwen2.5-0.5B-Instruct-GGUF 
  --include "Qwen2.5-0.5B-Instruct-Q4_K_M.gguf" --local-dir ~/llm-models

It will:

• Create the listing llm-models in your consumer’s dwelling folder
• Obtain the GGUF mannequin information safely into that folder. After obtain, you’ll be able to verify it’s there:

Output:

SmolLM2-135M-Instruct-Q4_K_M.gguf
Qwen2.5-0.5B-Instruct-Q4_K_M.gguf

// 3. Making a Llama-Swap Configuration

Llama-Swap makes use of a single YAML file to outline fashions and server instructions. Create a config.yaml file with contents like this:

fashions:
  "smollm2":
    cmd: |
      llama-server
      --model /path/to/fashions/llm-models/SmolLM2-135M-Instruct-Q4_K_M.gguf
      --port ${PORT}

  "qwen2.5":
    cmd: |
      llama-server
      --model /path/to/fashions/llm-models/Qwen2.5-0.5B-Instruct-Q4_K_M.gguf
      --port ${PORT}

Change /path/to/fashions/ along with your precise native path. Every entry underneath fashions: provides an ID (like "qwen2.5") and a shell cmd: to run its server. We use llama-server (from llama.cpp) with --model pointing to the GGUF file and --port ${PORT}. The ${PORT} macro tells Llama-Swap to assign a free port to every mannequin robotically. The teams part is optionally available. I’ve omitted it for this instance, so by default, Llama-Swap will solely run one mannequin at a time. You’ll be able to customise many choices per mannequin (aliases, timeouts, and many others.) on this configuration. For extra particulars on obtainable choices, see the Full Configuration Instance File.

// 4. Working Llama-Swap

With the binary and config.yaml prepared, begin Llama-Swap pointing to your config:

./llama-swap --config config.yaml --listen 127.0.0.1:8080

This launches the proxy server on localhost:8080. It’s going to learn config.yaml and (at first) load no fashions till the primary request arrives. Llama-Swap will now deal with API requests on port 8080, forwarding them to the suitable underlying llama-server course of primarily based on the "mannequin" parameter.

// 5. Interacting with Your Fashions

Now you may make OpenAI-style API calls to check every mannequin. Set up jq should you don’t have it earlier than operating the instructions under:

// Utilizing Qwen2.5

curl -s http://localhost:8080/v1/completions 
  -H "Content material-Kind: utility/json" 
  -H "Authorization: Bearer no-key" 
  -d '{
        "mannequin": "qwen2.5",
        "immediate": "Consumer: What's Python?nAssistant:",
        "max_tokens": 100
      }' | jq '.selections[0].textual content'

Output:
"Python is a well-liked general-purpose programming language. It's straightforward to study, has a big customary library, and is suitable with many working programs. Python is used for internet growth, information evaluation, scientific computing, and machine studying.nPython is a language that's fashionable for internet growth on account of its simplicity, versatility and its use of recent options. It's utilized in a variety of functions together with internet growth, information evaluation, scientific computing, machine studying and extra. Python is a well-liked language within the"

// Utilizing SmolLM2

curl -s http://localhost:8080/v1/completions 
  -H "Content material-Kind: utility/json" 
  -H "Authorization: Bearer no-key" 
  -d '{
        "mannequin": "smollm2",
        "immediate": "Consumer: What's Python?nAssistant:",
        "max_tokens": 100
      }' | jq '.selections[0].textual content'

Output:
"Python is a high-level programming language designed for simplicity and effectivity. It is recognized for its readability, syntax, and flexibility, making it a well-liked alternative for freshmen and builders alike.nnWhat is Python?"

Every mannequin will reply based on its coaching. The great thing about Llama-Swap is you don’t must restart something manually — simply change the "mannequin" discipline, and it handles the remainder. As proven within the examples above, you will see:

• qwen2.5: a extra verbose, technical response
• smollm2: an easier, extra concise reply

That confirms Llama-Swap is routing requests to the right mannequin!

# Conclusion

 
Congratulations! You have arrange Llama-Swap to run two LLMs on one machine, and now you can swap between them on the fly through API calls. We put in a proxy, ready a YAML configuration with two fashions, and noticed how Llama-Swap routes requests to the right backend.

Subsequent steps: You’ll be able to broaden this to incorporate:

• Bigger fashions (like TinyLlama, Phi-2, Mistral)
• Teams for concurrent serving
• Integration with LangChain, FastAPI, or different frontends

Have enjoyable exploring completely different fashions and configurations!
 
 

Kanwal Mehreen is a machine studying engineer and a technical author with a profound ardour for information science and the intersection of AI with drugs. She co-authored the e book “Maximizing Productiveness with ChatGPT”. As a Google Technology Scholar 2022 for APAC, she champions range and educational excellence. She’s additionally acknowledged as a Teradata Range in Tech Scholar, Mitacs Globalink Analysis Scholar, and Harvard WeCode Scholar. Kanwal is an ardent advocate for change, having based FEMCodes to empower ladies in STEM fields.