Products · Synology · NAS
Synology RS3626xs
- Processor
- Intel Xeon D-1700 series
- Memory (RAM)
- 8 or 48 GB
- Released
- 2026
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 Xeon D-1700 series | 8.0 GB | 5.2 GB | 43 GB/s | 153 of 622 | — |
| Intel Xeon D-1700 series · upgraded to the maker's maximum | 48.0 GB | 40.3 GB | 43 GB/s | 221 of 622 | — |
Specifications 8.0 GB configuration
| Maker | Synology |
|---|---|
| Type | NAS |
| Processor | Intel Xeon D-1700 series NAS (8 GB) |
| Memory (RAM) | 8.0 GB |
| Usable for AI models | 5.2 GB |
| Memory bandwidth | 43 GB/s |
| FP16 compute | 0.5 TFLOPS |
| Runs models on | CPU cores |
| Drive bays | 12x 3.5" SATA HDD or 2.5" SATA SSD |
| NAS system | DSM 7.x |
| Containers (Docker) | Yes |
| Virtual machines | Yes |
| Memory upgrade | DDR4 ECC UDIMM · 6 slots · up to 48.0 GB per the maker · ECC |
| Graphics card options | PCIe: 2 x Gen4 x8 slot (x8 link). Maker-supported graphics cards: NVIDIA RTX PRO 2000 Blackwell (16 GB); NVIDIA RTX 2000 Ada (16 GB); NVIDIA RTX PRO 4000 SFF Blackwell (24 GB); NVIDIA RTX 4000 SFF Ada (20 GB). |
| Notes | Intel Xeon D-1726. x86-64, up to 48 GB RAM, Container Manager and VMM, and it is one of six Synology systems that officially take an NVIDIA RTX card (16-24 GB VRAM). Synology itself runs local embedding/OCR/speech/caption models on that GPU. |
| Product page | www.synology.com (opens in a new tab) |
Running AI on the Synology RS3626xs
Yes, it runs AI models on the NAS itself. 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
| App | How it installs |
|---|---|
| Ollama + Open WebUI | In a container (Docker) |
| Any Linux guest running llama.cpp / Ollama | In a virtual machine |
AI features from Synology
| Feature | What it does | Where 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 |
| AI Console GPU-powered models (Synology Drive AI search) | With a supported NVIDIA card installed, AI Console runs local embedding, OCR, speech-to-text and image-captioning models (2-3 GB each) for semantic search in Synology Drive. Synology recommends reserving at least 16 GB of system memory. | On the NAS |
Best LLMs for the Synology RS3626xs 8.0 GB
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | Nemotron Labs Dory 1.0 4B | Q4_K_M | 3.9 GB | 10.6 tok/s | Runs well |
| 58Good quality (B) · 58 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | Gemma 4 E2B | Q4_K_M | 4.2 GB | 18.4 tok/s | Runs well |
| 50Fair quality (C) · 50 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking. | Ling 3.0 Tiny (8B-A1.3B) | Q3_K_M | 5.1 GB | 37.0 tok/s | Runs welltight fit |
| 62Good quality (B) · 62 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | Spark-X2.6 4B | Q4_K_M | 3.8 GB | 10.3 tok/s | Runs well |
| 64Good quality (B) · 64 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 75/100 (quality, fit, speed and memory use) sets the ranking. | Nanbeige4.2 3B | Q4_K_M | 5.0 GB | 10.1 tok/s | Runs welltight fit |
| 43Fair quality (C) · 43 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 74/100 (quality, fit, speed and memory use) sets the ranking. | Unlimited OCR 2 (3B-A0.6B) | Q6_K | 4.3 GB | 41.5 tok/s | Runs great |
| 58Good quality (B) · 58 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | Nemotron 3.5 Bolt E4B | Q3_K_M | 5.1 GB | 12.1 tok/s | Runs welltight fit |
| 57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | TMax 4B | Q4_K_M | 3.9 GB | 10.0 tok/s | Runs well |
| 57Good quality (B) · 57 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking. | MiniCPM5 2B | Q4_K_M | 2.9 GB | 16.8 tok/s | Runs well |
| 57Good quality (B) · 57 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking. | Gemma 4 E4B | Q3_K_M | 5.1 GB | 12.1 tok/s | Runs welltight fit |
Best Image models for the Synology RS3626xs 8.0 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 71/100 (quality, fit, speed and memory use) sets the ranking. | SDXL Turbo | FP8 | 4.2 GB | 11 s | Runs great |
| 72Great quality (A) · 72 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 70/100 (quality, fit, speed and memory use) sets the ranking. | FLUX.2 [klein] 4B | Q4 GGUF | 4.6 GB | 1.7 min | Slowtight fit · encoder unloaded after use |
| 90Excellent quality (A+) · 90 of 100: how capable the model is at Q8 GGUF, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking. | FLUX 3 Image Klein | Q8 GGUF | 4.3 GB | 11.8 min | Very slowCPU offload |
| 28Basic quality (D) · 28 of 100: how capable the model is at FP16, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | SD Turbo | FP16 | 3.3 GB | 6.6 s | Runs great |
| 82Great quality (A) · 82 of 100: how capable the model is at FP8, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking. | Z-Image Turbo | FP8 | 4.4 GB | 4.0 min | Very slowCPU offload |
| 78Great quality (A) · 78 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 65/100 (quality, fit, speed and memory use) sets the ranking. | FLUX.2 [klein] 9B | Q4 GGUF | 4.3 GB | 3.9 min | Very slowCPU offload |
Best Video models for the Synology RS3626xs 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 58/100 (quality, fit, speed and memory use) sets the ranking. | AnimateDiff v3 (SD 1.5) | FP16 | 4.7 GB | 27.6 min | Slowtight fit |
| 49Fair quality (C) · 49 of 100: how capable the model is at FP8, whatever it runs on. Match score 53/100 (quality, fit, speed and memory use) sets the ranking. | LTX-Video 2B | FP8 | 4.0 GB | 40.4 min | Very slowCPU offload |
| 44Fair quality (C) · 44 of 100: how capable the model is at FP8, whatever it runs on. Match score 47/100 (quality, fit, speed and memory use) sets the ranking. | CogVideoX 2B | FP8 | 4.7 GB | 1.6 h | Very slowtight fit · CPU offload |
Best Decision models for the Synology RS3626xs 8.0 GB
| Quality | Model | Quant | Memory | Speed | Runs |
|---|---|---|---|---|---|
| 78Great quality (A) · 78 of 100: how capable the model is at F16, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking. | OpenAI Privacy Filter | F16 | 3.8 GB | 59 ms | Runs great |
| 68Good quality (B) · 68 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | Llama Prompt Guard 2 22M | Q8_0 | 0.7 GB | 140 ms | Runs well |
| 86Excellent quality (A+) · 86 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 83/100 (quality, fit, speed and memory use) sets the ranking. | Jina Reranker v3.5 | Q8_0 | 1.9 GB | 1.1 s | Slow |
| 80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | GLiNER2.5 Decide | Q8_0 | 1.2 GB | 474 ms | Slow |
| 84Great quality (A) · 84 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 82/100 (quality, fit, speed and memory use) sets the ranking. | Jina Reranker v3 | Q8_0 | 1.9 GB | 1.1 s | Slow |
| 80Great quality (A) · 80 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 81/100 (quality, fit, speed and memory use) sets the ranking. | FRIDA-Decisions | Q8_0 | 1.7 GB | 791 ms | Slow |
| 87Excellent quality (A+) · 87 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. | Qwen3 Reranker 4B | Q5_K_M | 4.2 GB | 7.2 s | Very slow |
| 74Great quality (A) · 74 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking. | GLiClass ModernBERT Large v3 | Q8_0 | 1.1 GB | 388 ms | Runs well |
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.