Instructions to use murillo2000/Kimi-K3-0.40B-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 murillo2000/Kimi-K3-0.40B-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 murillo2000/Kimi-K3-0.40B-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf murillo2000/Kimi-K3-0.40B-GGUF:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf murillo2000/Kimi-K3-0.40B-GGUF:F16
# Run inference directly in the terminal:
llama cli -hf murillo2000/Kimi-K3-0.40B-GGUF:F16

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 murillo2000/Kimi-K3-0.40B-GGUF:F16
# Run inference directly in the terminal:
./llama-cli -hf murillo2000/Kimi-K3-0.40B-GGUF:F16

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 murillo2000/Kimi-K3-0.40B-GGUF:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf murillo2000/Kimi-K3-0.40B-GGUF:F16

Use Docker

docker model run hf.co/murillo2000/Kimi-K3-0.40B-GGUF:F16

LM Studio
Jan

vLLM

How to use murillo2000/Kimi-K3-0.40B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "murillo2000/Kimi-K3-0.40B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "murillo2000/Kimi-K3-0.40B-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/murillo2000/Kimi-K3-0.40B-GGUF:F16

Ollama
How to use murillo2000/Kimi-K3-0.40B-GGUF with Ollama:
```
ollama run hf.co/murillo2000/Kimi-K3-0.40B-GGUF:F16
```

Unsloth Studio

How to use murillo2000/Kimi-K3-0.40B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for murillo2000/Kimi-K3-0.40B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for murillo2000/Kimi-K3-0.40B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for murillo2000/Kimi-K3-0.40B-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use murillo2000/Kimi-K3-0.40B-GGUF with Docker Model Runner:
```
docker model run hf.co/murillo2000/Kimi-K3-0.40B-GGUF:F16
```

Lemonade

How to use murillo2000/Kimi-K3-0.40B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull murillo2000/Kimi-K3-0.40B-GGUF:F16

Run and chat with the model

lemonade run user.Kimi-K3-0.40B-GGUF-F16

List all available models

lemonade list

Kimi-K3 0.40B GGUF validation fixtures

GGUF conversions of the tiny Kimi-K3 0.40B development checkpoints.

These files are intended for:

Kimi-K3 architecture validation;
llama.cpp conversion and inference testing;
MXFP4 MoE layout testing;
out-of-core expert-loading research.

They are not intended as useful language models and should not be treated as quality proxies for the full Kimi-K3 model.

Files

`Kimi-K3-0.40B-F16.gguf`

Reference F16 conversion of:

inference-optimization/Kimi-K3-0.40B

`Kimi-K3-0.40B-MXFP4.gguf`

Hybrid conversion of:

inference-optimization/Kimi-K3-0.40B-MXFP4

Representation:

168 routed expert projection tensors remain in MXFP4;
they are repacked into 21 GGUF expert groups;
35 resident MoE tensors are dequantized to F16;
unknown packed tensors are rejected by the converter.

Converter

Converted with:

repository: murillo128/llama.cpp
branch: k3/out-of-core
converter commit: 84245db4c790af22135f34992689edcc11877003

See conversion-manifest.json for:

exact source revisions;
generated file sizes;
SHA-256 checksums;
conversion counts;
Python package versions.

Smoke-test prompt

According to all known laws

Both F16 and hybrid MXFP4 fixtures were loaded and executed successfully using the CPU backend. The observed fixture continuation was:

the start.

This is only a loading and execution smoke test, not a correctness or quality benchmark.

Project

These artifacts support the k3-out-of-core project, which explores tiered resident, RAM, and NVMe storage for routed MoE experts in llama.cpp.

Downloads last month: 580

GGUF

Model size

0.4B params

Architecture

kimi-k3

Hardware compatibility

16-bit

View +1 variant

Model tree for murillo2000/Kimi-K3-0.40B-GGUF

Base model

moonshotai/Kimi-K3

Finetuned

inference-optimization/Kimi-K3-0.40B

Quantized

(2)

this model