Synology DVA3221

Products · · NAS

Synology DVA3221

Processor
Intel Atom C3538 / C3558
Memory (RAM)
8 or 32 GB
Released
2020

Product page (opens in a new tab)Discontinued Full AI report

Configurations

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

Configuration Memory Usable for AI Bandwidth LLMs that run well Launch price
Intel Atom C3538 / C3558 8.0 GB 5.2 GB 19 GB/s 55 of 622 —
Intel Atom C3538 / C3558 · upgraded to the maker's maximum 32.0 GB 26.2 GB 19 GB/s 57 of 622 —

Specifications 8.0 GB configuration

MakerSynology
TypeNAS
ProcessorIntel Atom C3538 / C3558 NAS (8 GB)
Memory (RAM)8.0 GB
Usable for AI models5.2 GB
Memory bandwidth19 GB/s
FP16 compute0.1 TFLOPS
Runs models onCPU cores
Drive bays4x 3.5" SATA HDD or 2.5" SATA HDD/SSD
NAS systemDSM 7.x
Containers (Docker)Yes
Virtual machinesYes
Memory upgradeDDR4 non-ECC SODIMM · 2 slots · up to 32.0 GB per the maker · non-ECC
AvailabilityDiscontinued (sold used or refurbished)
NotesIntel Atom C3538. 4-core Atom C3538 CPU is slow for LLMs. It has a built-in 4 GB-class NVIDIA card (NVIDIA GeForce GTX 1650) that Synology uses for video analytics; Synology does not document using it from Docker, so treat LLM use as CPU-only unless you verify GPU passthrough yourself.
Product pagewww.synology.com (opens in a new tab)Discontinued

Running AI on the Synology DVA3221

It can, with limits: expect small models only, at modest speed. Models run on its CPU cores, and this configuration leaves about 5.2 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

AppHow it installs
Ollama + Open WebUI In a container (Docker)
Any Linux guest running llama.cpp / Ollama In a virtual machine

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

AI features from Synology

FeatureWhat it doesWhere the AI runs
Synology AI Console (cloud providers) Admin console that connects Synology Office, MailPlus, Chat, Drive etc. to Amazon Bedrock, Azure OpenAI, Baidu AI Cloud, Google AI Studio/Vertex AI, OpenAI or any OpenAI-compatible endpoint. Cloud service or another computer
AI Console de-identification Masks names and other sensitive entities before text is sent to a remote AI provider, using small local language models run through Container Manager. On the NAS
Surveillance Station Deep Video Analytics On-device computer-vision models (people/vehicle detection, counting, intrusion, face and license-plate recognition) run on the unit's own GPU. Vision only; not an LLM feature. On the NAS

Best LLMs for the Synology DVA3221 8.0 GB

Quality Model Quant Memory Speed Runs
58Good quality (B) · 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 71/100 (quality, fit, speed and memory use) sets the ranking. Gemma 4 E2BGoogle Q4_K_M 4.2 GB 4.3 tok/s Slow
57Good quality (B) · 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. Nemotron Labs Dory 1.0 4BNVIDIA Q3_K_M 3.4 GB 3.2 tok/s Slow
58Good quality (B) · 58 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. MiniCPM5 2BOpenBMB Q5_K_M 3.2 GB 3.2 tok/s Slow
55Good quality (B) · 55 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. Spark-X2.6 4BSparkLLM Q3_K_M 3.3 GB 3.1 tok/s Slow
57Good quality (B) · 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. Nanbeige4.2 3BBOSS Zhipin (Nanbeige) Q3_K_M 4.5 GB 3.0 tok/s Slowtight fit
41Fair quality (C) · 41 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. Unlimited OCR 2 (3B-A0.6B)Baidu Q4_K_M 3.6 GB 15.5 tok/s Runs well
42Fair quality (C) · 42 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. Ling 3.0 Tiny (8B-A1.3B)Ant Group (inclusionAI) Q2_K 4.4 GB 11.9 tok/s Runs well
51Fair quality (C) · 51 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. TMax 4BAi2 Q3_K_M 3.3 GB 3.0 tok/s Slow
54Fair quality (C) · 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. MiMo V2.6 Distill Qwen 9BXiaomi Q2_K 5.0 GB 1.7 tok/s Very slowtight fit
49Fair quality (C) · 49 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 66/100 (quality, fit, speed and memory use) sets the ranking. Nemotron 3.5 Bolt E4BNVIDIA Q2_K 4.4 GB 3.6 tok/s Slow

Best Image models for the Synology DVA3221 8.0 GB

Quality Model Quant Memory Speed Runs
90Excellent quality (A+) · 90 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking. FLUX 3 Image KleinBlack Forest Labs Q8 GGUF 4.3 GB 57.8 min Very slowCPU offload
72Great quality (A) · 72 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. FLUX.2 [klein] 4BBlack Forest Labs Q4 GGUF 4.6 GB 8.5 min Very slowtight fit · encoder unloaded after use
82Great quality (A) · 82 of 100: how capable the model is at FP8, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. Z-Image TurboAlibaba (Tongyi) FP8 4.4 GB 19.8 min Very slowCPU offload
81Great quality (A) · 81 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. ERNIE-ImageBaidu Q4 GGUF 4.1 GB 3.9 h Very slowCPU offload
81Great quality (A) · 81 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. MOSS-Image SFTOpenMOSS Q8 GGUF 4.1 GB 2.4 h Very slowCPU offload
39Basic quality (D) · 39 of 100: how capable the model is at FP8, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking. SDXL TurboStability AI FP8 4.2 GB 55 s Slow

Best Video models for the Synology DVA3221 8.0 GB

Quality Model Quant Memory Speed Runs
35Basic quality (D) · 35 of 100: how capable the model is at FP16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking. AnimateDiff v3 (SD 1.5)Shanghai AI Lab FP16 4.7 GB 2.3 h Very slowtight fit
49Fair quality (C) · 49 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. LTX-Video 2BLightricks FP8 4.0 GB 3.3 h Very slowCPU offload
44Fair quality (C) · 44 of 100: how capable the model is at FP8, whatever it runs on. Match score 46/100 (quality, fit, speed and memory use) sets the ranking. CogVideoX 2BZhipu AI FP8 4.7 GB 7.9 h Very slowtight fit · CPU offload

Best Decision models for the Synology DVA3221 8.0 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 1.9 GB 17.2 s Very slow
87Excellent quality (A+) · 87 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking. Qwen3 Reranker 4BAlibaba Q5_K_M 4.2 GB 114.8 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 1.9 GB 17.2 s Very slow
85Excellent quality (A+) · 85 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. zerank-2ZeroEntropy Q5_K_M 4.2 GB 114.8 s Very slow
85Excellent quality (A+) · 85 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking. Shieldstral 1.0 3BMistral AI Q6_K 4.4 GB 110.1 s Very slow
78Great quality (A) · 78 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. OpenAI Privacy FilterOpenAI Q8_0 2.4 GB 777 ms 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 7.6 s Very slow
82Great quality (A) · 82 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. mxbai-rerank-large v2Mixedbread Q8_0 2.7 GB 44.1 s Very slow

Other products with the Intel Atom C3538 / C3558 NAS (8 GB)

Spec sources

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.

Was this page helpful?