Instructions to use DevQuasar/moonshotai.Kimi-K3-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 DevQuasar/moonshotai.Kimi-K3-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 DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
# Run inference directly in the terminal:
llama cli -hf DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

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 DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
# Run inference directly in the terminal:
./llama-cli -hf DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

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 DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

Use Docker

docker model run hf.co/DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

LM Studio
Jan

vLLM

How to use DevQuasar/moonshotai.Kimi-K3-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DevQuasar/moonshotai.Kimi-K3-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DevQuasar/moonshotai.Kimi-K3-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

Ollama
How to use DevQuasar/moonshotai.Kimi-K3-GGUF with Ollama:
```
ollama run hf.co/DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
```
Unsloth Desktop
Docker Model Runner
How to use DevQuasar/moonshotai.Kimi-K3-GGUF with Docker Model Runner:
```
docker model run hf.co/DevQuasar/moonshotai.Kimi-K3-GGUF:BF16
```

Lemonade

How to use DevQuasar/moonshotai.Kimi-K3-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DevQuasar/moonshotai.Kimi-K3-GGUF:BF16

Run and chat with the model

lemonade run user.moonshotai.Kimi-K3-GGUF-BF16

List all available models

lemonade list

Atomic Chat

'Make knowledge free for everyone'

Experimental!

at this point i'm not suggeting this to be used for anyting else than test!

Based on https://github.com/ggml-org/llama.cpp/pull/26185

Q2_K | text-only | 962200.03 MiB (2.90 BPW)

Zeroshot demo with Q3_K_M Prompt: "write a html based realistic (graphics and physiscs and behavior) simulattion of flies in a jar"

html file

Quantized version of: moonshotai/Kimi-K3

Downloads last month: 328

GGUF

Model size

2.8T params

Architecture

kimi-k3

Hardware compatibility

2-bit

3-bit

16-bit

Model tree for DevQuasar/moonshotai.Kimi-K3-GGUF

Base model

moonshotai/Kimi-K3

Quantized

(55)

this model

Collection including DevQuasar/moonshotai.Kimi-K3-GGUF

Very Large GGUFs

Collection

GGUF quantized versions of very large models - over 100B parameters • 91 items • Updated Aug 26 • 10