Instructions to use prometheusAIR/Kimi-K3-REAP55-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 prometheusAIR/Kimi-K3-REAP55-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 prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
# Run inference directly in the terminal:
llama cli -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
# Run inference directly in the terminal:
llama cli -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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 prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
# Run inference directly in the terminal:
./llama-cli -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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 prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

Use Docker

docker model run hf.co/prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

LM Studio
Jan

vLLM

How to use prometheusAIR/Kimi-K3-REAP55-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "prometheusAIR/Kimi-K3-REAP55-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": "prometheusAIR/Kimi-K3-REAP55-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

Ollama
How to use prometheusAIR/Kimi-K3-REAP55-GGUF with Ollama:
```
ollama run hf.co/prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
```
Unsloth Desktop

How to use prometheusAIR/Kimi-K3-REAP55-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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": "prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use prometheusAIR/Kimi-K3-REAP55-GGUF with Docker Model Runner:
```
docker model run hf.co/prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M
```

Lemonade

How to use prometheusAIR/Kimi-K3-REAP55-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

Run and chat with the model

lemonade run user.Kimi-K3-REAP55-GGUF-IQ1_M

List all available models

lemonade list

Hermes Agent

How to use prometheusAIR/Kimi-K3-REAP55-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 prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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 prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use prometheusAIR/Kimi-K3-REAP55-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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 "prometheusAIR/Kimi-K3-REAP55-GGUF:IQ1_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"

Multilingual model request

by Pragmatism0220 - opened Jul 30

Discussion

Pragmatism0220

Jul 30

Thank you for your hard work! Considering users from different countries, would you consider developing a multilingual model? Based on the capabilities of Kimi K3, its coding proficiency, agent capabilities, tool-use features, and multilingual performance are truly outstanding.

prometheusAIR

Owner Jul 30

Yes, I usually aim to preserve full language capabilities in my models, but this is a pruning experiment - not for production use. For those use cases, I would strongly recommend using Kimi-K3 base version. It's already multilingual, and that contributes to its massive size.

Most users can't even load K3 on local hardware, and most people only use one language at a time. So one objective of this experiment was to answer: Can we prune K3 to a usable size for a single user by calibrating to their chosen language? Then we could make language-specific builds that people could actually run.

Unfortunately, this model is not what I would consider usable for those purposes, even in English. For now, it's an early data point for how K3 responds to aggressive pruning, and hopefully will lead to better attempts.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment