Minisforum N5 Air AI NAS

Configurations

Pick one to see which models it runs. More memory lets larger models fit; bandwidth sets how fast they answer.

Specifications 16.0 GB configuration

Maker	MMinisforum
Type	NAS
Processor	AMD Ryzen 7 8845HS / Ryzen 7 255 NAS (16 GB)
Memory (RAM)	16.0 GB
Usable for AI models	8.0 GB
Memory bandwidth	90 GB/s
FP16 compute	16.6 TFLOPS
Runs models on	GPU (Vulkan)
Drive bays	5x 3.5/2.5in SATA (30TB each) + 3x M.2 NVMe
NAS system	MinisCloud OS
Containers (Docker)	Yes
Virtual machines	Yes
Memory upgrade	DDR5 SO-DIMM 5600 MT/s (non-ECC) · 2 slots · up to 96.0 GB per the maker · non-ECC
Graphics card options	PCIe: 1x internal PCIe x16 slot, PCIe 4.0 x4 electrical. External GPU: 1x OCuLink port (rear); 2x USB4.
Notes	AMD Ryzen 7 255. 8-core Zen 4 with Radeon 780M, no NPU listed, up to 96GB dual-channel DDR5, Docker in MinisCloud OS or a replacement OS. Small models run at CPU speeds; a GPU in the PCIe slot or on OCuLink is needed for faster or larger models.
Product page	store.minisforum.com (opens in a new tab)

Running AI on the Minisforum N5 Air AI NAS

It can, with limits: expect small models only, at modest speed. Models run on its GPU (Vulkan), and this configuration leaves about 8.0 GB for them after the NAS system and its storage services. The grades below assume the NAS isn’t busy with backups or media work at the same time. More memory is the upgrade that matters most: see the configurations.

Apps you can install to run models

Apps that run LLMs on a NAS, with their container images

AI features from Minisforum

Best LLMs for the Minisforum N5 Air AI NAS 16.0 GB

Quality	Model	Quant	Memory	Speed	Runs
B66Good quality (B) · 66 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	Nemotron 3.5 Bolt E4BNVIDIA	Q4_K_M	6.3 GB	27.2 tok/s	Runs well
A76Great quality (A) · 76 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	IOK2 Horizon 7BInstitute of Foundation Models	Q4_K_M	7.9 GB	13.6 tok/s	Runs welltight fit
B64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	BZNanbeige4.2 3BBOSS Zhipin (Nanbeige)	Q4_K_M	5.3 GB	29.3 tok/s	Runs well
B64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Gemma 4 E4BGoogle	Q4_K_M	6.3 GB	27.2 tok/s	Runs well
B60Good quality (B) · 60 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	Gemma 4 E2BGoogle	Q5_K_M	4.9 GB	43.9 tok/s	Runs great
A74Great quality (A) · 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	XMiMo V2.6 Distill Qwen 9BXiaomi	Q4_K_M	7.2 GB	13.0 tok/s	Runs welltight fit
A74Great quality (A) · 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	STSenseNova Loop 9B PreviewSenseTime	Q4_K_M	7.6 GB	12.7 tok/s	Runs welltight fit
B59Good quality (B) · 59 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	AGLing 3.0 Tiny (8B-A1.3B)Ant Group (inclusionAI)	Q6_K	7.9 GB	52.8 tok/s	Runs welltight fit
B58Good quality (B) · 58 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	JBMellum2.1 Thinking (12B-A2.5B)JetBrains	Q3_K_M	7.3 GB	52.5 tok/s	Runs welltight fit
A73Great quality (A) · 73 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	DROrnith 1.5 9BDeepReinforce	Q4_K_M	7.4 GB	12.7 tok/s	Runs welltight fit

Best Image models for the Minisforum N5 Air AI NAS 16.0 GB

Quality	Model	Quant	Memory	Speed	Runs
A77Great quality (A) · 77 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	BFFLUX.2 [klein] 4BBlack Forest Labs	Q8 GGUF	6.4 GB	6.0 s	Runs greatencoder unloaded after use
A78Great quality (A) · 78 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking.	Z-Image TurboAlibaba (Tongyi)	Q4 GGUF	6.1 GB	16 s	Runs wellencoder unloaded after use
A84Great quality (A) · 84 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking.	BFFLUX 3 Image KleinBlack Forest Labs	Q4 GGUF	5.7 GB	36 s	Runs wellencoder unloaded after use
B56Good quality (B) · 56 of 100: how capable the model is at FP8, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	LDreamShaper XL v2 TurboLykon	FP8	5.6 GB	5.8 s	Runs great
A79Great quality (A) · 79 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	BERNIE-Image TurboBaidu	Q4 GGUF	7.4 GB	20 s	Runs welltight fit · encoder unloaded after use
C52Fair quality (C) · 52 of 100: how capable the model is at FP8, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	BDSDXL-Lightning (4-step)ByteDance	FP8	5.6 GB	3.1 s	Runs great

Best Video models for the Minisforum N5 Air AI NAS 16.0 GB

Quality	Model	Quant	Memory	Speed	Runs
C49Fair quality (C) · 49 of 100: how capable the model is at FP8, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	LLTX-Video 2BLightricks	FP8	6.1 GB	2.0 min	Runs greatencoder unloaded after use
D35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.	SAAnimateDiff v3 (SD 1.5)Shanghai AI Lab	FP16	4.7 GB	1.7 min	Runs great
C44Fair quality (C) · 44 of 100: how capable the model is at FP8, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking.	ZACogVideoX 2BZhipu AI	FP8	6.2 GB	4.7 min	Runs wellencoder unloaded after use
C53Fair quality (C) · 53 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	Wan 2.1 T2V 1.3BAlibaba	FP8	7.1 GB	18.9 min	Slowtight fit · encoder unloaded after use
C53Fair quality (C) · 53 of 100: how capable the model is at FP8, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	ZACogVideoX 5BZhipu AI	FP8	5.9 GB	17.5 min	SlowCPU offload

Best Decision models for the Minisforum N5 Air AI NAS 16.0 GB

Quality	Model	Quant	Memory	Speed	Runs
A+86Excellent quality (A+) · 86 of 100: how capable the model is at F16, whatever it runs on. Match score 93/100 (quality, fit, speed and memory use) sets the ranking.	JAJina Reranker v3.5Jina AI	F16	2.7 GB	48 ms	Runs great
A84Great quality (A) · 84 of 100: how capable the model is at F16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking.	JAJina Reranker v3Jina AI	F16	2.7 GB	48 ms	Runs great
A82Great quality (A) · 82 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	Mmxbai-rerank-large v2Mixedbread	Q8_0	2.9 GB	88 ms	Runs great
A81Great quality (A) · 81 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	LAd1 3BLiquid AI	Q6_K	3.8 GB	101 ms	Runs great
A80Great quality (A) · 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	SAFRIDA-DecisionsSber AI	F16	2.4 GB	24 ms	Runs great
A80Great quality (A) · 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	FGLiNER2.5 DecideFastino	F16	1.7 GB	16 ms	Runs great
A80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	AGLemonDecider 4BAnt Group (inclusionAI)	Q8_0	6.5 GB	99 ms	Runs great
A80Great quality (A) · 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3 Reranker 0.6BAlibaba	F16	2.7 GB	48 ms	Runs great

Other products with the AMD Ryzen 7 8845HS / Ryzen 7 255 NAS (16 GB)

• N11 9-bay NAS motherboard (AMD Ryzen 7 8845HS)TTopton · NAS · 8 or 16 GB · $510Product page (opens in a new tab)
• N31 10GbE NAS motherboard (Ryzen 7 8845HS / 7840U / PRO 6850U / PRO 6650U)TTopton · NAS · 8 or 16 GB · $327Product page (opens in a new tab)
• N5 NASMMinisforum · NAS · 16 or 96 GB · $599Product page (opens in a new tab)Discontinued