LLMs · Naive AI

Naive N0.5 Flash (309B-A15.5B)

Name: Naive N0.5 Flash (309B-A15.5B)
Author: Naive AI

Parameters: 309B (15.5B active)
Architecture: Mixture of experts
Max context: 1,024K tokens
License: mit
Released: 2026-09

How to run it

Original weights · Hugging Face

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

huggingface.co/NaiveAI/Naive-N0.5-Flash (opens in a new tab)

No sign-in needed.

Official repo · GitHub

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

github.com/NaiveAI-Labs/Naive-N0.5-Flash (opens in a new tab)

git clone https://github.com/NaiveAI-Labs/Naive-N0.5-Flash

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

Apps that run Naive N0.5 Flash (309B-A15.5B)

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)

Docker Model Runner

Handy if you already work in Docker.

Also good:

Text Generation WebUI

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
vLLM	Server	Original weights	WSLWindowsLinux	vLLM	`vllm serve NaiveAI/Naive-N0.5-Flash`
SGLang	Server	Original weights	WSLWindowsLinux	SGLang	`python3 -m sglang.launch_server --model-path NaiveAI/Naive-N0.5-Flash`
Docker Model Runner	CLI + API	Original weights	macOSWindowsLinux	llama.cpp, vLLM	`docker model run hf.co/NaiveAI/Naive-N0.5-Flash`
Text Generation WebUI	Desktop app	Original weights	macOSWindowsLinux	llama.cpp, Transformers, ExLlamaV3	Get Text Generation WebUI

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: sliding-window hybrid: 9 of 48 layers are global (DeepSeek Sparse Attention, full KV cache kept); the other 39 use a 128-token window.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	124 GB	124 GB	125 GB	All files (opens in a new tab)
`Q3_K_M`	Low	151 GB	152 GB	153 GB	All files (opens in a new tab)
`Q4_K_M`	Good	188 GB	189 GB	190 GB	All files (opens in a new tab)
`Q5_K_M`	High	219 GB	220 GB	221 GB	All files (opens in a new tab)
`Q6_K`	High	253 GB	254 GB	255 GB	All files (opens in a new tab)
`Q8_0`	Very high	327 GB	328 GB	329 GB	All files (opens in a new tab)
`F16`	Full	618 GB	619 GB	619 GB	Original weights (opens in a new tab)

Best hardware for Naive N0.5 Flash (309B-A15.5B)

Quality	Hardware	Quant	Memory	Speed	Runs
92Excellent quality (A+) · 92 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking.	Apple M3 Ultra (512 GB)	`Q5_K_M`	251 GB / 384 GB	39.7 tok/s	Runs great
79Great quality (A) · 79 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	Apple M3 Ultra (256 GB)	`Q3_K_M`	168 GB / 192 GB	58.5 tok/s	Runs welltight fit
66Good quality (B) · 66 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking.	Apple M2 Ultra (192 GB)	`Q2_K`	136 GB / 144 GB	70.0 tok/s	Runs welltight fit
94Excellent quality (A+) · 94 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking.	Apple M5 Ultra (512 GB)	`Q6_K`	285 GB / 384 GB	47.1 tok/s	Runs great
92Excellent quality (A+) · 92 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI350X	`Q5_K_M`	235 GB / 288 GB	175 tok/s	Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI325X	`Q4_K_M`	202 GB / 256 GB	172 tok/s	Runs great
90Excellent quality (A+) · 90 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 95/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip	`Q4_K_M`	202 GB / 252 GB	194 tok/s	Runs great
79Great quality (A) · 79 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 90/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI300X	`Q3_K_M`	162 GB / 192 GB	186 tok/s	Runs great
79Great quality (A) · 79 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	Apple M5 Ultra (256 GB)	`Q3_K_M`	168 GB / 192 GB	78.8 tok/s	Runs welltight fit
79Great quality (A) · 79 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA B200	`Q3_K_M`	161 GB / 180 GB	224 tok/s	Runs welltight fit

Check it on your own hardware

Run it on a cluster

Devices that can’t run Naive N0.5 Flash (309B-A15.5B) alone can run it together. These example clusters do:

Try it with your own devices

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.