LLMs · KRAFTON

Raon-1 mini Preview (30B-A4B)

Name: Raon-1 mini Preview (30B-A4B)
Author: KRAFTON

Parameters: 29B (4.3B active)
Architecture: Mixture of experts
Max context: 256K tokens
License: apache-2.0
Released: 2026-10

How to run it

Original weights · Hugging Face

Full-precision files from KRAFTON, for fine-tuning, vLLM, Transformers.

huggingface.co/KRAFTON/Raon-1-mini-preview (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from KRAFTON. The weights themselves download from Hugging Face.

github.com/krafton-ai/Raon-1-mini-preview (opens in a new tab)

KRAFTON's official repo. KRAFTON also publishes GGUF files (KRAFTON/Raon-1-mini-preview-gguf), but for now they need KRAFTON's llama.cpp fork (krafton-ai/llama.cpp) because upstream llama.cpp support is still pending, so LM Studio and Ollama cannot load them yet. An official MLX version is at KRAFTON/Raon-1-mini-preview-mlx.

git clone https://github.com/krafton-ai/Raon-1-mini-preview

Links checked 5 days ago against Hugging Face, Ollama and GitHub.

Apps that run Raon-1 mini Preview (30B-A4B)

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

LM Studio

Runs the MLX build, typically 20–35% faster than GGUF for a model with 4.3B active parameters on Apple silicon.

Also good:

Windows PC, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Text Generation WebUI

Windows PC, AMD Radeon

Docker Model Runner

Handy if you already work in Docker.

Also good:

Text Generation WebUI

Linux, NVIDIA GPU

Docker Model Runner

Handy if you already work in Docker.

Also good:

Text Generation WebUI

Linux server, 80 GB+ GPUs

vLLM

The full model fits in GPU memory at 8-bit, so it can batch many requests: best for serving a team or an app.

Also good:

App	Type	Opens it as	Systems	Engine	Start
LM Studio	Desktop app	MLX	macOSWindowsLinux	llama.cpp, MLX	Open in LM Studio
Jan	Desktop app	MLX	macOSWindowsLinux	llama.cpp, MLX	Get Jan
MLX LM	CLI + API	MLX	macOS	MLX	`mlx_lm.chat --model KRAFTON/Raon-1-mini-preview-mlx`
vLLM	Server	Original weights	WSLWindowsLinux	vLLM	`vllm serve KRAFTON/Raon-1-mini-preview`
SGLang	Server	Original weights	WSLWindowsLinux	SGLang	`python3 -m sglang.launch_server --model-path KRAFTON/Raon-1-mini-preview`
Docker Model Runner	CLI + API	Original weights	macOSWindowsLinux	llama.cpp, vLLM	`docker model run hf.co/KRAFTON/Raon-1-mini-preview`
Unsloth Desktop	Desktop app	MLX	macOSWindowsLinux	llama.cpp, MLX, stable-diffusion.cpp	Get Unsloth Desktop
Atomic Chat	Desktop app	MLX	macOSWindowsLinux	llama.cpp, MLX-VLM	Open in Atomic Chat
Pi	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp or MLX LM	Point it at a llama.cpp or MLX LM server
Hermes Agent	Assistant	MLX	macOSWSLWindowsLinux	uses llama.cpp or MLX LM	Point it at a llama.cpp or MLX LM server
OpenClaw	Assistant	MLX	macOSWSLWindowsLinux	uses llama.cpp or MLX LM	Point it at a llama.cpp or MLX LM server
OpenCode	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Codex CLI	Coding agent	MLX	macOSWSLWindowsLinux	uses Ollama or any OpenAI-compatible server	Point it at a llama.cpp or MLX LM server
OpenHands	Coding agent	MLX	macOSWSLWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Zed	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Cline	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Open Interpreter	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Goose	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Aider	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Crush	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Qwen Code	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Kilo Code	Coding agent	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Open WebUI	Assistant	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Khoj	Assistant	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Letta	Assistant	MLX	macOSWindowsLinux	uses llama.cpp, Ollama or LM Studio	Point it at a llama.cpp or MLX LM server
Text Generation WebUI	Desktop app	Original weights	macOSWindowsLinux	llama.cpp, Transformers, ExLlamaV3	Get Text Generation WebUI
Msty Studio	Desktop app	MLX	macOSWindowsLinux	Ollama, llama.cpp, MLX	Get Msty Studio
Inferencer	Desktop app	MLX	macOS	MLX	Get Inferencer
exo	Server	MLX	macOSLinux	MLX	Get exo
Continue	Coding agent	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server
SillyTavern	Assistant	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server
LibreChat	Assistant	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server
Cherry Studio	Assistant	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server
Chatbox	Assistant	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server
PrivateGPT	Assistant	MLX	macOSWindowsLinux	uses Ollama, LM Studio or llama.cpp	Point it at a llama.cpp or MLX LM server

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.

KV cache: hybrid linear attention: 7 of 25 layers keep a latent (MLA) cache; the rest use Raon Delta Attention with a fixed-size state.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	11.5 GB	12.2 GB	12.4 GB	All files (opens in a new tab)
`Q3_K_M`	Low	14.1 GB	14.8 GB	15.0 GB	All files (opens in a new tab)
`Q4_K_M`	Good	17.6 GB	18.2 GB	18.4 GB	All files (opens in a new tab)
`Q5_K_M`	High	20.4 GB	21.1 GB	21.3 GB	All files (opens in a new tab)
`Q6_K`	High	23.6 GB	24.3 GB	24.5 GB	All files (opens in a new tab)
`Q8_0`	Very high	30.5 GB	31.2 GB	31.4 GB	All files (opens in a new tab)
`F16`	Full	57.6 GB	58.2 GB	58.4 GB	Original weights (opens in a new tab)

Best hardware for Raon-1 mini Preview (30B-A4B)

The least expensive GPU in our catalog that runs it without offloading is the AMD Radeon RX 7600 XT (launch price $329) at Q3_K_M.

Quality	Hardware	Quant	Memory	Speed	Runs
88Excellent quality (A+) · 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	Apple M1 Max (64 GB)	`Q8_0`	35.0 GB / 48.0 GB	42.0 tok/s	Runs great
88Excellent quality (A+) · 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	Apple M4 Max (64 GB)	`Q8_0`	35.0 GB / 48.0 GB	54.2 tok/s	Runs great
88Excellent quality (A+) · 88 of 100: how capable the model is at F16, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell	`F16`	63.0 GB / 96.0 GB	78.7 tok/s	Runs great
87Excellent quality (A+) · 87 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5090	`Q6_K`	25.9 GB / 32.0 GB	156 tok/s	Runs great
87Excellent quality (A+) · 87 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 94/100 (quality, fit, speed and memory use) sets the ranking.	Apple M4 Pro (48 GB)	`Q6_K`	27.1 GB / 36.0 GB	39.9 tok/s	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 7900 XTX	`Q4_K_M`	19.4 GB / 24.0 GB	132 tok/s	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 3090	`Q4_K_M`	19.4 GB / 24.0 GB	141 tok/s	Runs great
83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4090	`Q4_K_M`	19.4 GB / 24.0 GB	146 tok/s	Runs great
88Excellent quality (A+) · 88 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking.	Apple M4 Max (128 GB)	`Q8_0`	38.9 GB / 96.0 GB	54.2 tok/s	Runs great
73Great quality (A) · 73 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	AMD Radeon RX 9070 XT	`Q3_K_M`	15.6 GB / 16.0 GB	123 tok/s	Runs welltight fit

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.