What AI models can the Marvell OCTEON TX2 CN9130 NAS (4 GB) run?

RAM: 4.0 GB
Usable for models: 2.0 GB
Memory bandwidth: 19 GB/s
Runs models on: CPU cores
Released: 2019

82 / 622 LLMs run Best: LFM3 1.7B 6 / 61 Image models run Best: SDXL Turbo 0 / 43 Video models run None fit 36 / 70 Decision models run Best: Jina Reranker v3.5

A NAS with this chip runs models on its processor cores, so memory speed sets the pace: small models (about 1-4B) answer at a usable speed, bigger ones are slow. We plan around 2.0 GB for models, what is left after the NAS system and its storage services. Models run in a container (Docker) or the maker’s app store, and speeds assume the NAS isn’t busy with storage work.

More about this hardware

Maker	Marvell
Type	NAS (network attached storage) processor
Share of memory for models	50% (what the NAS system and its storage services typically leave free at this size)
Chip	Marvell OCTEON TX2 CN9130
AI accelerator	None. Models run on the processor cores.
NAS models with this chip	QNAP TS-435XeU
Memory type	System RAM, shared with the NAS operating system and its storage services
Bandwidth figure	19.2 GB/s, estimated, see the note below
Note	One profile for every NAS built on this chip. Memory speed is an estimate from the chip’s supported memory in a typical NAS fitting; a unit with one memory module runs slower than one with two. Speed assumes the NAS is not busy with storage work.
Specifications	www.qnap.com

Apps that run models well on the Marvell OCTEON TX2 CN9130 NAS (4 GB)

Linux on ARM boards

For a typical model that fits in memory. Each model page picks the best app for that model on this hardware. All apps

1 product with the Marvell OCTEON TX2 CN9130 NAS (4 GB)

TS-435XeUQNAP · NAS · 4 or 32 GBProduct page (opens in a new tab)

Best LLMs for the Marvell OCTEON TX2 CN9130 NAS (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
53Fair quality (C) · 53 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.	LFM3 1.7BLiquid AI	`Q4_K_M`	1.9 GB	4.7 tok/s	Slowtight fit
35Basic quality (D) · 35 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	PaddleOCR-VL 1.6 0.9BBaidu	`Q3_K_M`	1.4 GB	11.0 tok/s	Runs well
42Fair quality (C) · 42 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	LFM3-VL 2BLiquid AI	`Q3_K_M`	1.8 GB	5.5 tok/s	Slowtight fit
38Basic quality (D) · 38 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM-V 4.7 1BOpenBMB	`Q4_K_M`	1.8 GB	5.3 tok/s	Slowtight fit
29Basic quality (D) · 29 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 60/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM-OCR 1.0 1.3BOpenBMB	`Q2_K`	1.4 GB	10.3 tok/s	Runs well
33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.	TMax 2BAi2	`Q3_K_M`	1.8 GB	5.6 tok/s	Slowtight fit
25Basic quality (D) · 25 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.	MiniCPM5 1BOpenBMB	`Q2_K`	1.4 GB	12.4 tok/s	Runs well
33Basic quality (D) · 33 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 2BAlibaba	`Q3_K_M`	2.0 GB	4.7 tok/s	Slowtight fit
30Basic quality (D) · 30 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.	LFM2.5 2.6BLiquid AI	`Q2_K`	2.0 GB	5.0 tok/s	Slowtight fit
22Basic quality (D) · 22 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 58/100 (quality, fit, speed and memory use) sets the ranking.	LFM2.5 1.2B InstructLiquid AI	`Q2_K`	1.3 GB	11.5 tok/s	Runs well
21Basic quality (D) · 21 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3.5 0.8BAlibaba	`Q3_K_M`	1.3 GB	12.2 tok/s	Runs well
29Basic quality (D) · 29 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 57/100 (quality, fit, speed and memory use) sets the ranking.	Jamba2 3BAI21 Labs	`Q2_K`	2.0 GB	4.4 tok/s	Slowtight fit

Best Image models for the Marvell OCTEON TX2 CN9130 NAS (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
39Basic quality (D) · 39 of 100: how capable the model is at FP8, whatever it runs on. Match score 50/100 (quality, fit, speed and memory use) sets the ranking.	SDXL TurboStability AI	`FP8`	1.6 GB	73 s	SlowCPU offload
42Fair quality (C) · 42 of 100: how capable the model is at FP16, whatever it runs on. Match score 47/100 (quality, fit, speed and memory use) sets the ranking.	Realistic Vision V5.1SG161222	`FP16`	1.3 GB	10.8 min	Very slowCPU offload
28Basic quality (D) · 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 46/100 (quality, fit, speed and memory use) sets the ranking.	SD TurboStability AI	`FP16`	1.3 GB	43 s	SlowCPU offload
40Fair quality (C) · 40 of 100: how capable the model is at FP16, whatever it runs on. Match score 46/100 (quality, fit, speed and memory use) sets the ranking.	DreamShaper 8Lykon	`FP16`	1.3 GB	10.8 min	Very slowCPU offload
35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 44/100 (quality, fit, speed and memory use) sets the ranking.	Stable Diffusion 1.5Stability AI / Runway	`FP16`	1.3 GB	10.8 min	Very slowCPU offload
30Basic quality (D) · 30 of 100: how capable the model is at FP16, whatever it runs on. Match score 44/100 (quality, fit, speed and memory use) sets the ranking.	LCM DreamShaper v7Tsinghua (LCM)	`FP16`	1.3 GB	2.2 min	Very slowCPU offload

Best Video models for the Marvell OCTEON TX2 CN9130 NAS (4 GB)

None of the video models in our catalog fit in this much memory.

Best Decision models for the Marvell OCTEON TX2 CN9130 NAS (4 GB)

Quality	Model	Quant	Memory	Speed	Runs
86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	Jina Reranker v3.5Jina AI	`Q8_0`	1.6 GB	24.7 s	Very slow
84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	Jina Reranker v3Jina AI	`Q8_0`	1.6 GB	24.7 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	GLiNER2.5 DecideFastino	`Q8_0`	1.2 GB	10.9 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.	FRIDA-DecisionsSber AI	`Q8_0`	1.7 GB	18.3 s	Very slow
80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	Qwen3 Reranker 0.6BAlibaba	`Q8_0`	1.6 GB	24.7 s	Very slow
76Great quality (A) · 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.	Llama Prompt Guard 2 86MMeta	`Q8_0`	1.0 GB	12.5 s	Very slow
79Great quality (A) · 79 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.	mxbai-rerank-large v2Mixedbread	`Q5_K_M`	1.9 GB	63.3 s	Very slowtight fit
76Great quality (A) · 76 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking.	mxbai-rerank-base v2Mixedbread	`Q8_0`	1.3 GB	20.2 s	Very slow

Best LLMs by task on this hardware Open in Model Finder

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.